[{"model":"alibaba-qwen4-flash","air":"alibaba:qwen@4-flash","name":"Qwen4-Flash","creator":"alibaba","endpoints":["chat-completions"],"status":"coming-soon","releasedAt":"2026-11-16T00:00:00Z","weight":5,"headline":"Lightweight Qwen4 model for fast, high-volume coding, office, and agent work","description":"Qwen4-Flash is the lightweight API model in Alibaba's Qwen4 family, built for fast, high-volume work. It carries the generation's gains in world knowledge, coding, and office productivity into its smallest API tier. It suits coding assistants, document and spreadsheet automation, and agent steps that run often and need quick responses.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen4-flash.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen4-flash.png"},{"model":"alibaba-qwen4-max","air":"alibaba:qwen@4-max","name":"Qwen4-Max","creator":"alibaba","endpoints":["chat-completions"],"status":"coming-soon","releasedAt":"2026-11-16T00:00:00Z","weight":5,"headline":"Flagship multimodal Qwen4 model for coding, office work, front-end design, and complex agent tasks","description":"Qwen4-Max is the flagship multimodal model in Alibaba's Qwen4 family, built for coding, office productivity, front-end design, and complex agent tasks. It builds on Qwen3.8-Max with stronger world knowledge, perception, and professional tool use, and it accepts text, images, and video with a 1M-token context window, configurable reasoning effort, function calling, and JSON Schema output. It is well suited to coding agents, screenshot and document analysis, and agents that operate graphical interfaces.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen4-max.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen4-max.png","contextWindow":1048576,"maxOutputTokens":393216},{"model":"alibaba-qwen4-plus","air":"alibaba:qwen@4-plus","name":"Qwen4-Plus","creator":"alibaba","endpoints":["chat-completions"],"status":"coming-soon","releasedAt":"2026-11-16T00:00:00Z","weight":5,"headline":"Mid-tier Qwen4 model for coding, office productivity, and agent workflows","description":"Qwen4-Plus is the mid-tier API model in Alibaba's Qwen4 family, between Qwen4-Max and Qwen4-Flash. It brings the generation's gains in world knowledge, perception, coding, office productivity, and front-end design to production assistants and agents, including agents that operate graphical interfaces.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen4-plus.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen4-plus.png"},{"model":"google-gemini-4-hydrogen","air":"google:gemini@4-hydrogen","name":"Gemini 4 Hydrogen","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-11-12T00:00:00Z","weight":5,"headline":"Google's fast, efficient Gemini 4 model for everyday coding, agent workflows, and knowledge work","description":"Gemini 4 Hydrogen is the fast, efficient model in Google's Gemini 4 family, built for everyday coding, agent workflows, and knowledge work. It accepts text, images, video, audio, and PDFs, with a 1M-token context window and configurable thinking effort. It is well suited to high-volume coding and agent tasks, worker and verifier agents in multi-model workflows, and turning long documents, spreadsheets, and recordings into reports and structured deliverables.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-4-hydrogen.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-4-hydrogen.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.0000015,"unit":"inputToken"},{"amount":0.0000075,"unit":"outputToken"},{"amount":7.5e-8,"unit":"cachedInputToken"}]},{"model":"google-gemini-4-carbon","air":"google:gemini@4-carbon","name":"Gemini 4 Carbon","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-11-10T00:00:00Z","weight":5,"headline":"Google's most capable Gemini model for the hardest coding, knowledge work, and scientific reasoning","description":"Gemini 4 Carbon is Google's most capable Gemini model, built for the hardest tasks in software engineering, enterprise knowledge work, and scientific and mathematical reasoning. It spends more tokens than Gemini 4 Argon to explore problems, reason deeply, and verify its own work, and it accepts text, images, video, audio, and PDFs with a 1M-token context window, up to 65,536 output tokens, and configurable thinking effort. It is well suited to multi-package code migrations, document synthesis for legal and financial analysis, research problems, and agent workflows that run extended build-and-verify loops.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-4-carbon.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-4-carbon.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"Through Nov 30, 2026"},{"amount":0.000004,"unit":"inputToken","label":"From Dec 1, 2026"},{"amount":0.00001,"unit":"outputToken","label":"Through Nov 30, 2026"},{"amount":0.00002,"unit":"outputToken","label":"From Dec 1, 2026"},{"amount":1e-7,"unit":"cachedInputToken","label":"Through Nov 30, 2026"},{"amount":2e-7,"unit":"cachedInputToken","label":"From Dec 1, 2026"}]},{"model":"minimax-h3-image","air":"minimax:h3@image","name":"MiniMax H3 Image","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-11-10T00:00:00Z","weight":5,"headline":"Text-to-image generation and multi-reference editing at up to 20 megapixels","description":"MiniMax H3 Image is an open-weights image generation and editing model from MiniMax. It creates images from text or edits and combines up to 16 reference images, at 1K, 2K or 4K, or at a custom width and height up to 20 megapixels. It follows counts, positions and layouts closely and renders long, dense text, which suits product and marketing visuals, posters and infographics, concept art, and compositions that merge several subjects.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-image.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-image.png","pricingBasis":"token","pricingRates":[{"amount":0.0000023,"unit":"inputToken","label":"Text input"},{"amount":0.0000015,"unit":"inputToken","label":"Image input"},{"amount":0.0000038,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0176","amount":0.017645,"latencyMs":15558,"exampleId":"championship-pigeon-racer-portrait"},{"configuration":"text-to-image","price":"$0.0189","amount":0.018865,"latencyMs":16323,"exampleId":"japanese-joinery-workshop-wall-chart"},{"configuration":"reference-to-image","price":"$0.0432","amount":0.043187,"latencyMs":26711,"exampleId":"planetarium-restoration-fundraising-panorama"},{"configuration":"reference-to-image","price":"$0.0432","amount":0.043232,"latencyMs":31258,"exampleId":"underground-mushroom-grower-editorial-portrait"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write text-to-image prompts for MiniMax H3 Image: sizing by preset, exact dimensions or no size, and planning the task and layout before the details.","date":"2026-11-10","url":"https://runware.ai/docs/models/minimax-h3-image/guides/prompting"},{"slug":"prompt-optimizer","title":"Prompt optimizer","description":"How to use the prompt optimizer on MiniMax H3 Image: what the rewrite adds to a short prompt, and when to turn it off and send the prompt as written.","date":"2026-11-10","url":"https://runware.ai/docs/models/minimax-h3-image/guides/prompt-optimizer"},{"slug":"text-and-layouts","title":"Text and layouts","description":"How to get exact copy and structured layouts from MiniMax H3 Image: quoted text, charts and figures, dense pages, app screens, and grids and storyboards.","date":"2026-11-10","url":"https://runware.ai/docs/models/minimax-h3-image/guides/text-and-layouts"},{"slug":"editing-images","title":"Editing images","description":"How to edit one image with MiniMax H3 Image: naming the target in words, moving, copying, replacing and deleting objects, rewriting text, and editing over several rounds.","date":"2026-11-10","url":"https://runware.ai/docs/models/minimax-h3-image/guides/editing-images"},{"slug":"multi-reference-composition","title":"Multi-reference composition","description":"How to compose one image from up to 16 references with MiniMax H3 Image: giving each image a role, borrowing color, following a layout sketch and furnishing a room.","date":"2026-11-10","url":"https://runware.ai/docs/models/minimax-h3-image/guides/multi-reference-composition"}]},{"model":"minimax-h3-image-highspeed","air":"minimax:h3@image-highspeed","name":"MiniMax H3 Image Highspeed","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-11-10T00:00:00Z","weight":5,"headline":"Faster H3 Image generation and multi-reference editing at up to 20 megapixels","description":"MiniMax H3 Image Highspeed is the faster variant of MiniMax H3 Image, with the same parameters, limits and output sizes. It creates images from text or edits and combines up to 16 reference images, at 1K, 2K or 4K, or at a custom width and height up to 20 megapixels. It suits interactive tools, high-volume pipelines and iterative editing where turnaround matters.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-image-highspeed.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-image-highspeed.png","pricingBasis":"token","pricingRates":[{"amount":0.0000023,"unit":"inputToken","label":"Text input"},{"amount":0.000003,"unit":"inputToken","label":"Image input"},{"amount":0.0000076,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0418","amount":0.041755,"latencyMs":14925,"exampleId":"botanical-garden-picture-book-spread"},{"configuration":"text-to-image","price":"$0.0422","amount":0.04219,"latencyMs":12728,"exampleId":"mycelium-materials-magazine-cover"},{"configuration":"reference-to-image","price":"$0.0725","amount":0.072454,"latencyMs":19802,"exampleId":"pneumatic-mail-history-book-cover"},{"configuration":"reference-to-image","price":"$0.0743","amount":0.074259,"latencyMs":21206,"exampleId":"historic-storefront-restoration-visualization"}]},{"model":"anthropic-claude-fable-5-2","air":"anthropic:claude@fable-5.2","name":"Claude Fable 5.2","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-11-09T00:00:00Z","weight":5,"headline":"Frontier multimodal model for ambitious, open-ended engineering, science, and agent work","description":"Claude Fable 5.2 is the latest release in Anthropic’s Fable line, built for the hardest open-ended technical work in software engineering, science, and research. It reads text and images, calls tools, and reasons with configurable effort, improving on Fable 5.1 in coding, reasoning, vision, and computer use while generating output faster. Its 1M-token context window and 128K-token maximum output suit long agent runs, large-codebase migrations, and coordinating teams of smaller models on complex projects.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-fable-5-2.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-fable-5-2.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000008,"unit":"inputToken"},{"amount":0.00004,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"},{"amount":0.00001,"unit":"cacheWriteToken"}]},{"model":"bytedance-seedream-5-5","air":"bytedance:seedream@5.5","name":"Seedream 5.5","creator":"bytedance","status":"live","releasedAt":"2026-11-09T00:00:00Z","weight":5,"headline":"Realistic image generation and consistent multi-turn editing, with native transparency and layer decomposition","description":"Seedream 5.5 is ByteDance's image generation and editing model, focused on visual realism, aesthetics and editing consistency. It supports text-to-image and reference-guided editing with up to 10 input images, keeps unedited areas stable across multi-turn edits, and outputs transparent images from a prompt or from references. It can also split an image into a base image and up to 99 editable layers, returning the text, position, font and style of each text layer, which suits poster, packaging and marketing design workflows.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-5-5.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-5-5.png","pricingRates":[{"amount":0.02,"unit":"output","label":"Up to 1.5K (2.61 MP)"},{"amount":0.035,"unit":"output","label":"2K (above 2.61 MP)"},{"amount":0.008,"unit":"inputImage"}],"pricingExamples":[{"configuration":"transparent-background · 2048×2048","price":"$0.035","amount":0.035,"latencyMs":20245,"exampleId":"beet-kvass-highball-menu-cutout"},{"configuration":"text-to-image · 2K","price":"$0.035","amount":0.035,"latencyMs":33427,"exampleId":"museum-fossil-lab-social-post"},{"configuration":"reference-to-image · 1280×1600","price":"$0.036","amount":0.036,"latencyMs":20101,"exampleId":"vintage-diving-helmet-catalog-cleanup"},{"configuration":"layer-decomposition","price":"$0.208","amount":0.208,"latencyMs":41458,"exampleId":"victorian-cable-drama-series-poster"},{"configuration":"layer-decomposition-ocr","price":"$0.208","amount":0.208,"latencyMs":50531,"exampleId":"pigeon-racing-documentary-streaming-poster"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write text-to-image prompts for Seedream 5.5: choosing between resolution tiers and exact sizes, layering a prompt into directives, and setting exact copy on the image.","date":"2026-11-09","url":"https://runware.ai/docs/models/bytedance-seedream-5-5/guides/prompting"},{"slug":"editing-images","title":"Editing images","description":"How to edit images with Seedream 5.5: an instruction to recolor, clean up or relight one photo, a drawn mark to pick the target, and edits chained over several rounds.","date":"2026-11-09","url":"https://runware.ai/docs/models/bytedance-seedream-5-5/guides/editing-images"},{"slug":"multi-reference-composition","title":"Multi-reference composition","description":"How to compose one image from up to ten references with Seedream 5.5: assigning roles, dressing one model in a garment range, arranging a product set and sizing the output.","date":"2026-11-09","url":"https://runware.ai/docs/models/bytedance-seedream-5-5/guides/multi-reference-composition"},{"slug":"layer-decomposition","title":"Layer decomposition","description":"How to split one image into a base image and transparent layers with Seedream 5.5, from layer positions and split size to reading the text on text layers.","date":"2026-11-09","url":"https://runware.ai/docs/models/bytedance-seedream-5-5/guides/layer-decomposition"},{"slug":"transparent-backgrounds","title":"Transparent backgrounds","description":"How to make cut-out assets with Seedream 5.5: transparent output from a prompt or an existing photo, the auto default, matching sticker sets and soft hair edges.","date":"2026-11-09","url":"https://runware.ai/docs/models/bytedance-seedream-5-5/guides/transparent-backgrounds"}]},{"model":"google-gemini-omni-flash-1-2","air":"google:gemini@omni-flash-1.2","name":"Gemini Omni Flash 1.2","creator":"google","zdr":true,"status":"live","releasedAt":"2026-11-05T00:00:00Z","weight":5,"headline":"Video generation and editing with native audio, 15-second clips, and output from 360p drafts to 4K","description":"Gemini Omni Flash 1.2 is Google's multimodal video model for generating, editing, and extending clips with native synchronized audio. It generates up to 15 seconds from text or a single starting image, edits source videos of up to 15 seconds with optional reference images, and keeps characters, products, and brand details consistent across reference images and videos. Output runs from 360p for quick drafts to 1080p and 4K for delivery, which suits high-volume work such as product clips, virtual try-ons, and ad variations.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/google-gemini-omni-flash-1-2.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-omni-flash-1-2.png","pricingRates":[{"amount":0.02,"unit":"durationSecond","label":"360p"},{"amount":0.07,"unit":"durationSecond","label":"720p"},{"amount":0.1,"unit":"durationSecond","label":"1080p"},{"amount":0.21,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"video-edit · 720p","price":"$0.423","amount":0.422613,"latencyMs":92284,"exampleId":"alpine-railway-blizzard-safety-shot"},{"configuration":"first-frame · 720p · 8s","price":"$0.56","amount":0.56,"latencyMs":61446,"exampleId":"cassette-shell-colorway-reveal"},{"configuration":"reference-to-video · 1080p · 6s","price":"$0.60","amount":0.6,"latencyMs":129358,"exampleId":"pangolin-release-day-for-night-relight"},{"configuration":"video-extend · 720p · 10s","price":"$0.70","amount":0.7,"latencyMs":99225,"exampleId":"couture-hearing-device-showcase"},{"configuration":"first-last-frame · 720p · 10s","price":"$0.70","amount":0.7,"latencyMs":76704,"exampleId":"tidal-barrier-nightfall-relight"},{"configuration":"text-to-video · 720p · 15s","price":"$1.05","amount":1.05,"latencyMs":79161,"exampleId":"vineyard-tractor-colorway-reveal-film"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write text-to-video prompts for Gemini Omni Flash 1.2: scene structure, one shot or several, timing, audio direction, and on-screen text.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/prompting"},{"slug":"resolution-and-duration","title":"Resolution and duration","description":"How to size and time clips with Gemini Omni Flash 1.2: resolution tiers from 360p to 4K, 16:9 and 9:16 pairs, cost per second, and the duration range of each mode.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/resolution-and-duration"},{"slug":"first-and-last-frame","title":"First and last frame","description":"How to animate an opening frame or connect a first and last frame with Gemini Omni Flash 1.2, from building a matching pair to describing the motion between them.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/first-and-last-frame"},{"slug":"reference-driven-video","title":"Reference images and videos","description":"How to build new clips from reference images and videos with Gemini Omni Flash 1.2, carrying a person, a product, a location or an illustration style into the shot.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/reference-driven-video"},{"slug":"replacing-with-reference-images","title":"Replacing elements with reference images","description":"How to edit a clip with Gemini Omni Flash 1.2 and a reference image: swap a garment for virtual try-on, swap a product for ad variations, or add furniture to a room.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/replacing-with-reference-images"},{"slug":"editing-video","title":"Editing video","description":"How to edit a clip with Gemini Omni Flash 1.2 from a text prompt: relight a scene, change the season or style, rewrite a sign, remove or add objects, and chain edits.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/editing-video"},{"slug":"extending-video","title":"Extending video","description":"How to extend a clip with Gemini Omni Flash 1.2: writing what happens next, keeping or cutting the shot, carrying the audio, and chaining extensions into a longer video.","date":"2026-11-05","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-2/guides/extending-video"}]},{"model":"openai-gpt-image-2-6-flare","air":"openai:gpt-image@2.6-flare","name":"GPT-Image-2.6 Flare","creator":"openai","status":"live","releasedAt":"2026-11-05T00:00:00Z","weight":5,"headline":"Fast image generation and editing for interactive iteration","description":"GPT-Image-2.6 Flare is OpenAI's fast image model for generation and editing in interactive workflows. Besides raster images, it generates SVG graphics and animated GIFs from text, and it edits images from marks drawn over them, like a circle around the object to change. A thinking level sets how much it plans before rendering, separate from rendering quality.","capabilities":["io:text-to-image","io:image-to-image","op:edit","op:vectorize","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-2-6-flare.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-2-6-flare.png","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.00003,"unit":"outputToken","label":"Image output"},{"amount":7.5e-7,"unit":"inputToken","label":"SVG authoring · GPT-5.4 Mini · Input"},{"amount":0.0000045,"unit":"outputToken","label":"SVG authoring · GPT-5.4 Mini · Output"},{"amount":0.0000025,"unit":"inputToken","label":"SVG authoring · GPT-5.4 · Input"},{"amount":0.000015,"unit":"outputToken","label":"SVG authoring · GPT-5.4 · Output"},{"amount":0.000005,"unit":"inputToken","label":"SVG authoring · GPT-5.5 · Input"},{"amount":0.00003,"unit":"outputToken","label":"SVG authoring · GPT-5.5 · Output"},{"amount":0.000002,"unit":"inputToken","label":"SVG authoring · GPT-5.6 Terra · Input"},{"amount":0.000012,"unit":"outputToken","label":"SVG authoring · GPT-5.6 Terra · Output"},{"amount":1e-7,"unit":"inputToken","label":"SVG authoring · GPT-6 Luna · Input"},{"amount":5e-7,"unit":"outputToken","label":"SVG authoring · GPT-6 Luna · Output"},{"amount":0.00001,"unit":"inputToken","label":"SVG authoring · GPT-6 Astra · Input"},{"amount":0.00005,"unit":"outputToken","label":"SVG authoring · GPT-6 Astra · Output"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.0587","amount":0.058655,"latencyMs":14109,"exampleId":"mycology-benefit-supper-invitation"},{"configuration":"text-to-gif","price":"$0.118","amount":0.11831,"latencyMs":12493,"exampleId":"analog-seismograph-blog-hero"},{"configuration":"text-to-image","price":"$0.167","amount":0.16685,"latencyMs":29720,"exampleId":"luthier-workshop-team-composite"},{"configuration":"text-to-vector","price":"$0.174","amount":0.174045,"latencyMs":37015,"exampleId":"urban-cooling-campaign-key-visual"},{"configuration":"transparent-background","price":"$0.175","amount":0.17479,"latencyMs":14034,"exampleId":"subsea-cable-robot-blog-hero"},{"configuration":"reference-to-image","price":"$0.193","amount":0.192609,"latencyMs":19492,"exampleId":"electric-ferry-launch-invitation"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt GPT-Image-2.6 Flare: writing a brief it can plan a layout from, rendering readable text in the scene, and choosing a size from its dimension envelope.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/prompting"},{"slug":"quality-and-thinking-level","title":"Quality and thinking level","description":"How to set the two effort controls on GPT-Image-2.6 Flare: settings.thinkingLevel to plan a layout, settings.quality to render it, and when to lower each one.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/quality-and-thinking-level"},{"slug":"editing-images","title":"Editing images","description":"How to edit images with GPT-Image-2.6 Flare: one source and a written instruction to remove, recolor, add or replace, and several references combined into one image.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/editing-images"},{"slug":"animated-gifs","title":"Animated GIFs","description":"How to generate looping animated GIFs with GPT-Image-2.6 Flare: prompting for motion, setting frame count and timing with settings.gif, and sizing the frames.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/animated-gifs"},{"slug":"markup-editing","title":"Markup editing","description":"How to point GPT-Image-2.6 Flare at what to edit with a markup layer: building it, circling one item among look-alikes and marking where a new object goes.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/markup-editing"},{"slug":"svg-output","title":"SVG output","description":"How to generate editable SVG files with GPT-Image-2.6 Flare: the reference image behind each file, prompting for clean shapes, and picking the author with settings.svg.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/svg-output"},{"slug":"transparent-backgrounds","title":"Transparent backgrounds","description":"How to generate cut-out assets with GPT-Image-2.6 Flare: settings.background and the formats it needs, prompts that keep edges clean, contact shadows and frosted glass.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-flare/guides/transparent-backgrounds"}]},{"model":"openai-gpt-image-2-6-sunburst","air":"openai:gpt-image@2.6-sunburst","name":"GPT-Image-2.6 Sunburst","creator":"openai","status":"live","releasedAt":"2026-11-05T00:00:00Z","weight":5,"headline":"Image generation and editing for detailed compositions and precise edits","description":"GPT-Image-2.6 Sunburst is OpenAI's image model for detailed compositions and editing work that needs precise control. Besides raster images, it generates SVG graphics and animated GIFs from text, and it edits images from marks drawn over them, like a circle around the object to change. A thinking level sets how much it plans before rendering, separate from rendering quality.","capabilities":["io:text-to-image","io:image-to-image","op:edit","op:vectorize","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-2-6-sunburst.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-2-6-sunburst.png","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.00003,"unit":"outputToken","label":"Image output"},{"amount":7.5e-7,"unit":"inputToken","label":"SVG authoring · GPT-5.4 Mini · Input"},{"amount":0.0000045,"unit":"outputToken","label":"SVG authoring · GPT-5.4 Mini · Output"},{"amount":0.0000025,"unit":"inputToken","label":"SVG authoring · GPT-5.4 · Input"},{"amount":0.000015,"unit":"outputToken","label":"SVG authoring · GPT-5.4 · Output"},{"amount":0.000005,"unit":"inputToken","label":"SVG authoring · GPT-5.5 · Input"},{"amount":0.00003,"unit":"outputToken","label":"SVG authoring · GPT-5.5 · Output"},{"amount":0.000002,"unit":"inputToken","label":"SVG authoring · GPT-5.6 Terra · Input"},{"amount":0.000012,"unit":"outputToken","label":"SVG authoring · GPT-5.6 Terra · Output"},{"amount":1e-7,"unit":"inputToken","label":"SVG authoring · GPT-6 Luna · Input"},{"amount":5e-7,"unit":"outputToken","label":"SVG authoring · GPT-6 Luna · Output"},{"amount":0.00001,"unit":"inputToken","label":"SVG authoring · GPT-6 Astra · Input"},{"amount":0.00005,"unit":"outputToken","label":"SVG authoring · GPT-6 Astra · Output"}],"pricingExamples":[{"configuration":"text-to-gif","price":"$0.0374","amount":0.037375,"latencyMs":31112,"exampleId":"geothermal-plant-director-looping-portrait"},{"configuration":"text-to-image","price":"$0.0431","amount":0.04313,"latencyMs":61404,"exampleId":"subsea-cable-landing-wall-art"},{"configuration":"text-to-vector","price":"$0.0634","amount":0.063389,"latencyMs":59669,"exampleId":"nocturnal-moth-survey-festival-poster"},{"configuration":"inpaint","price":"$0.0771","amount":0.077143,"latencyMs":57407,"exampleId":"cave-rescue-equipment-sticker-pack"},{"configuration":"transparent-background","price":"$0.0788","amount":0.07884,"latencyMs":32340,"exampleId":"oyster-farm-childrens-comic-panel"},{"configuration":"reference-to-image","price":"$0.192","amount":0.191921,"latencyMs":111116,"exampleId":"adaptive-rainwear-fashion-lookbook"}],"guides":[{"slug":"detailed-editing","title":"Detailed editing","description":"How to choose GPT-Image-2.6 Sunburst over Flare for dense, exact text and precise edits: a side-by-side comparison, a chain of focused edits and a markup edit.","date":"2026-11-05","url":"https://runware.ai/docs/models/openai-gpt-image-2-6-sunburst/guides/detailed-editing"}]},{"model":"prunaai-p-video-2-pro-replace","air":"prunaai:p-video@2-pro-replace","name":"P-Video-2-Pro-Replace","creator":"prunaai","status":"live","releasedAt":"2026-11-05T00:00:00Z","weight":5,"headline":"Replace one or two people in an existing video from a photo, keeping motion, camera, and lip sync","description":"P-Video-2-Pro-Replace is Pruna AI's character replacement model that puts the person from one or two photos into an existing video. It keeps the original camera, background, timing, and motion, carries over the source soundtrack with lips that follow it, and replaces two people at once with a prompt that says who replaces whom. It takes source clips from 1 second to 2 minutes and outputs from 480p up to 4K, for UGC ad variations, localized spokesperson videos, and recasting existing footage.","capabilities":["io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-2-pro-replace.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-2-pro-replace.png","pricingRates":[{"amount":0.05,"unit":"durationSecond","label":"480p"},{"amount":0.1,"unit":"durationSecond","label":"768p"},{"amount":0.12,"unit":"durationSecond","label":"1080p"},{"amount":0.14,"unit":"durationSecond","label":"2K"},{"amount":0.18,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"two-person-replacement","price":"$0.35","amount":0.35,"latencyMs":102125,"exampleId":"period-wharf-crew-cleanup"},{"configuration":"two-person-replacement","price":"$0.35","amount":0.35,"latencyMs":88156,"exampleId":"museum-training-cast-continuity-fix"},{"configuration":"video-edit · 1080p","price":"$0.42","amount":0.42,"latencyMs":121023,"exampleId":"community-observatory-logo-bumper"},{"configuration":"video-edit · 1080p","price":"$0.42","amount":0.42,"latencyMs":148070,"exampleId":"mushroom-farm-sensor-explainer"}],"guides":[{"slug":"replacing-people-in-video","title":"Replacing people in video","description":"How to put the person from a photo into an existing video with Pruna P-Video-2-Pro-Replace, keeping the camera, the motion and the soundtrack of the source.","date":"2026-11-05","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro-replace/guides/replacing-people-in-video"},{"slug":"choosing-the-photo","title":"Choosing the photo","description":"How to choose the photo for Pruna P-Video-2-Pro-Replace, which sets both the person and the outfit: a headshot, a full look, a kept uniform or a 3D character.","date":"2026-11-05","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro-replace/guides/choosing-the-photo"},{"slug":"lip-sync-and-soundtrack","title":"Lip sync and soundtrack","description":"How to work with the source audio in Pruna P-Video-2-Pro-Replace: lip sync, the original voice, voiceover clips and outputs with no audio track.","date":"2026-11-05","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro-replace/guides/lip-sync-and-soundtrack"},{"slug":"mapping-photos-to-people","title":"Mapping photos to people","description":"How to tell Pruna P-Video-2-Pro-Replace who to replace: the left-to-right default with two photos, a reversed mapping, and one person singled out in a group.","date":"2026-11-05","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro-replace/guides/mapping-photos-to-people"}]},{"model":"moonshotai-kimi-k3-1-pro","air":"moonshotai:kimi@k3.1-pro","name":"Kimi K3.1 Pro","creator":"moonshot-ai","zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-11-04T00:00:00Z","weight":5,"headline":"Flagship open-weight reasoning LLM for agentic coding, 3D creation, and financial research","description":"Kimi K3.1 Pro is Moonshot AI's flagship open-weight model, a post-trained successor to Kimi K3 built for coding, agent workflows, and knowledge-intensive work. It pairs a 1M-token context window with always-on reasoning at configurable effort, image understanding, structured outputs, and tool calling, and improves on K3 in coding, 3D modeling, financial analysis, and factual accuracy. It is well suited to coding agents that build full projects and games, scripted 3D and CAD modeling in tools such as Blender, and research assistants that work through financial documents and spreadsheets.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/moonshotai-kimi-k3-1-pro.jpg","ogImage":"https://assets.runware.ai/og/moonshotai-kimi-k3-1-pro.png","contextWindow":1048576,"maxOutputTokens":1048576,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken"},{"amount":0.00001,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"}]},{"model":"klingai-video-4-0-flash","air":"klingai:kling-video@4.0-flash","name":"Kling VIDEO 4.0 Flash","creator":"klingai","status":"live","releasedAt":"2026-11-04T00:00:00Z","weight":5,"headline":"720p video generation and editing from text, images, and references, with clips up to 20 seconds","description":"Kling VIDEO 4.0 Flash is the 720p model of the Kling 4.0 family, generating clips of up to 20 seconds from text, a first-frame image, reference images, reference videos, and reusable elements. It combines several references in a single request to hold subjects, style, and motion, and it edits an existing video from a prompt. It suits social content, previews of longer productions, and other work where 720p output is enough.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-4-0-flash.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-4-0-flash.png","pricingRates":[{"amount":0.094,"unit":"durationSecond","label":"720p · No video input"},{"amount":0.126,"unit":"durationSecond","label":"720p · Video input"}],"pricingExamples":[{"configuration":"video-edit","price":"$0.647","amount":0.6475,"latencyMs":67249,"exampleId":"fossil-archive-dance-teaser"},{"configuration":"first-frame · 720p","price":"$0.799","amount":0.799,"latencyMs":1002010,"exampleId":"alpine-shelter-weather-resilience-film"},{"configuration":"text-to-video · 720p","price":"$0.799","amount":0.799,"latencyMs":95517,"exampleId":"mushroom-farm-music-video-segment"},{"configuration":"elements · 1464×628","price":"$0.806","amount":0.806,"latencyMs":74896,"exampleId":"tactile-map-workshop-b-roll"},{"configuration":"reference-to-video · 960×960","price":"$1.07","amount":1.071,"latencyMs":90997,"exampleId":"nocturnal-rooftop-apiary-relight"}]},{"model":"topazlabs-wonder-4","air":"topazlabs:wonder@4","name":"Topaz Labs Wonder 4","creator":"topazlabs","zdr":true,"status":"live","releasedAt":"2026-11-03T00:00:00Z","weight":5,"headline":"All-in-one generative image enhancement that restores, denoises, and upscales in a single pass","description":"Wonder 4 is Topaz Labs' all-purpose generative model for image enhancement and upscaling. It treats each region of an image separately, combining denoising, deblurring, compression artifact removal, face recovery, and upscaling in one pass with no settings to tune. It suits photo restoration, cleanup of degraded or compressed images, and enhancement at the original size or enlarged up to six times, with output up to 256 megapixels.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/topazlabs-wonder-4.jpg","ogImage":"https://assets.runware.ai/og/topazlabs-wonder-4.png","pricingRates":[{"amount":0.0428,"unit":"output","label":"1 MP output"},{"amount":0.0428,"unit":"output","label":"4 MP output"},{"amount":0.0428,"unit":"output","label":"8 MP output"},{"amount":0.0856,"unit":"output","label":"16 MP output"},{"amount":0.1284,"unit":"output","label":"24 MP output"},{"amount":0.1712,"unit":"output","label":"32 MP output"},{"amount":0.214,"unit":"output","label":"40 MP output"},{"amount":0.2996,"unit":"output","label":"50 MP output"},{"amount":0.3424,"unit":"output","label":"64 MP output"},{"amount":0.5564,"unit":"output","label":"100 MP output"},{"amount":0.6848,"unit":"output","label":"128 MP output"},{"amount":1.3696,"unit":"output","label":"256 MP output"}],"pricingExamples":[{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":23476,"exampleId":"heritage-seed-catalog-linocut"},{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":26344,"exampleId":"entomology-specimen-archive-image-cleanup"},{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":19924,"exampleId":"deep-sea-exhibit-panorama-enlargement"},{"configuration":"upscale","price":"$0.0856","amount":0.0856,"latencyMs":32503,"exampleId":"pipe-organ-conservator-portrait"}],"guides":[{"slug":"upscaling-images","title":"Upscaling images","description":"How to upscale images with Topaz Wonder 4: sending the request, choosing a factor from 2 to 6, and keeping the source inside the output size limits.","date":"2026-11-03","url":"https://runware.ai/docs/models/topazlabs-wonder-4/guides/upscaling-images"},{"slug":"enhancing-at-original-resolution","title":"Enhancing at the original resolution","description":"How to clean up an image with Topaz Wonder 4 without enlarging it: what a factor of 1 returns, when to pick it over 2, and how it fits fixed-size listings.","date":"2026-11-03","url":"https://runware.ai/docs/models/topazlabs-wonder-4/guides/enhancing-at-original-resolution"},{"slug":"text-and-graphics","title":"Text and graphics","description":"How to upscale packaging, dashboards and printed menus with Topaz Wonder 4, so every line of copy the source shows comes back sharp and spelled the same.","date":"2026-11-03","url":"https://runware.ai/docs/models/topazlabs-wonder-4/guides/text-and-graphics"},{"slug":"transparent-images","title":"Transparent images","description":"How to upscale transparent PNGs with Topaz Wonder 4 so product cutouts, see-through glass and loose hair keep their transparency, and why a JPG output loses it.","date":"2026-11-03","url":"https://runware.ai/docs/models/topazlabs-wonder-4/guides/transparent-images"},{"slug":"restoring-photos","title":"Restoring photos","description":"How to restore damaged photos with Topaz Wonder 4: faded print scans, over-compressed uploads, low-light noise and motion blur, each fixed in one pass.","date":"2026-11-03","url":"https://runware.ai/docs/models/topazlabs-wonder-4/guides/restoring-photos"}]},{"model":"minimax-h3-highspeed","air":"minimax:h3@highspeed","name":"MiniMax H3 Highspeed","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-10-30T00:00:00Z","weight":5,"headline":"Fast H3 video generation for everyday content, with first-to-last frame control and multimodal references","description":"MiniMax H3 Highspeed is a member of the MiniMax H3 family that balances generation quality and speed. It generates 4 to 15 second videos at 480p or 768p from text, from a first frame or a first and last frame pair, or from reference images, videos and audio. It is well suited to everyday content creation, social video and production workflows where turnaround matters.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-highspeed.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-highspeed.png","pricingRates":[{"amount":0.04,"unit":"durationSecond","label":"480p"},{"amount":0.06,"unit":"durationSecond","label":"768p"},{"amount":0.04,"unit":"durationSecond","label":"480p · uploaded video"},{"amount":0.06,"unit":"durationSecond","label":"768p · uploaded video"},{"amount":0.03,"unit":"inputImage","after":5,"label":"each image beyond the first five"}],"pricingExamples":[{"configuration":"text-to-video · 1344×768 · 5s","price":"$0.30","amount":0.3,"latencyMs":33204,"exampleId":"industrial-flamenco-music-video-segment"},{"configuration":"first-last-frame · 480p · 8s","price":"$0.32","amount":0.32,"latencyMs":25248,"exampleId":"palimpsest-documentary-title-sequence"},{"configuration":"first-frame · 768p · 6s","price":"$0.36","amount":0.36,"latencyMs":75205,"exampleId":"automatic-flood-barrier-product-demo"},{"configuration":"reference-to-video · 1024×768 · 10s","price":"$0.96","amount":0.96,"latencyMs":49128,"exampleId":"botanical-crime-series-teaser"}]},{"model":"minimax-h3-lite","air":"minimax:h3@lite","name":"MiniMax H3 Lite","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-10-30T00:00:00Z","weight":5,"headline":"Lightweight H3 video generation for previews, drafts and batch variations, with frame and reference control","description":"MiniMax H3 Lite is the lightweight member of the MiniMax H3 family, built for quick, low-cost iteration. It generates 4 to 15 second videos at 480p or 768p from text, from a first frame or a first and last frame pair, or from reference images, videos and audio. It is well suited to concept previews, creative drafts and batch variations that explore a direction before a final render.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-lite.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-lite.png","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"480p"},{"amount":0.04,"unit":"durationSecond","label":"768p"},{"amount":0.025,"unit":"durationSecond","label":"480p · uploaded video"},{"amount":0.04,"unit":"durationSecond","label":"768p · uploaded video"},{"amount":0.02,"unit":"inputImage","after":5,"label":"each image beyond the first five"}],"pricingExamples":[{"configuration":"text-to-video · 1344×768 · 5s","price":"$0.20","amount":0.2,"latencyMs":22698,"exampleId":"subsea-cable-documentary-title-teaser"},{"configuration":"first-frame · 768p · 6s","price":"$0.24","amount":0.24,"latencyMs":34569,"exampleId":"seed-vault-curator-presenter-clip"},{"configuration":"first-last-frame · 768p · 8s","price":"$0.32","amount":0.32,"latencyMs":46476,"exampleId":"apiary-thriller-title-sequence"},{"configuration":"reference-to-video · 480p · 8s","price":"$0.35","amount":0.35,"latencyMs":68526,"exampleId":"satellite-risk-analyst-briefing-clip"}]},{"model":"prunaai-p-video-upscale","air":"prunaai:p-video@upscale","name":"P-Video-Upscale","creator":"prunaai","status":"live","releasedAt":"2026-10-29T00:00:00Z","weight":5,"headline":"Video upscaling to 720p, 1080p, 2K, or 4K that keeps the source aspect ratio and audio","description":"P-Video-Upscale is Pruna AI's video upscaling model for raising an existing clip to 720p, 1080p, 2K, or 4K while keeping its aspect ratio and audio track. It sharpens what is already in the frame without a prompt, can raise the frame rate to as much as 60 FPS by interpolating new frames, and offers a turbo mode that trades a little quality for speed. It works on live action, anime, game footage, and archival film, for bringing low-resolution generations or older footage up to delivery resolution.","capabilities":["io:video-to-video","op:upscale","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-upscale.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-upscale.png","pricingRates":[{"amount":0.01,"unit":"durationSecond","label":"720p · Up to 30 fps"},{"amount":0.02,"unit":"durationSecond","label":"720p · Above 30 fps"},{"amount":0.02,"unit":"durationSecond","label":"1080p · Up to 30 fps"},{"amount":0.04,"unit":"durationSecond","label":"1080p · Above 30 fps"},{"amount":0.04,"unit":"durationSecond","label":"2K · Up to 30 fps"},{"amount":0.08,"unit":"durationSecond","label":"2K · Above 30 fps"},{"amount":0.08,"unit":"durationSecond","label":"4K · Up to 30 fps"},{"amount":0.16,"unit":"durationSecond","label":"4K · Above 30 fps"}],"pricingExamples":[{"configuration":"upscale","price":"$0.242","amount":0.242,"latencyMs":73508,"exampleId":"archival-kintsugi-workshop-restoration"},{"configuration":"upscale","price":"$0.242","amount":0.242,"latencyMs":92890,"exampleId":"peregrine-falcon-conservation-portrait"},{"configuration":"upscale","price":"$0.242","amount":0.242,"latencyMs":73060,"exampleId":"gyotaku-artwork-detail-enhancement"},{"configuration":"upscale","price":"$0.242","amount":0.242,"latencyMs":90000,"exampleId":"archival-lace-mill-film-restoration"}],"guides":[{"slug":"upscaling-video","title":"Upscaling video","description":"How to upscale video with Pruna P-Video-Upscale: the request, the size each preset returns, the frame cap on clip length, the audio track and turbo mode.","date":"2026-10-29","url":"https://runware.ai/docs/models/prunaai-p-video-upscale/guides/upscaling-video"},{"slug":"raising-the-frame-rate","title":"Raising the frame rate","description":"How to raise a clip's frame rate with Pruna P-Video-Upscale: interpolating to 30 or 60 fps, picking a rate, and fitting the price and length limits above 30.","date":"2026-10-29","url":"https://runware.ai/docs/models/prunaai-p-video-upscale/guides/raising-the-frame-rate"},{"slug":"upscaling-generated-video","title":"Upscaling generated video","description":"How to finish P-Video-2-Pro clips at 1080p, 2K or 4K with Pruna P-Video-Upscale: chaining by UUID, fitting the 4K frame cap, and choosing the render tier.","date":"2026-10-29","url":"https://runware.ai/docs/models/prunaai-p-video-upscale/guides/upscaling-generated-video"}]},{"model":"openai-gpt-6-1-terra","air":"openai:gpt@6.1-terra","name":"GPT-6.1 Terra","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-10-26T00:00:00Z","weight":5,"headline":"Fast, efficient model for visual reasoning, computer use, and high-volume professional work","description":"GPT-6.1 Terra is OpenAI's fast, efficient model for visual reasoning, computer use, and everyday professional work. It improves on GPT-5.6 Terra in interpreting images and technical drawings, operating desktop applications, coding, and factual accuracy, and combines a 1,050,000-token context window with up to 128,000 output tokens, image understanding, structured outputs, function calling, and broad tool support. Configurable reasoning from none through max suits high-volume automation such as form filling and data entry, document drafting, and agents that act on application screens.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-6-1-terra.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-6-1-terra.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000001,"unit":"inputToken","label":"≤ 272K context"},{"amount":0.000002,"unit":"inputToken","label":"> 272K context"},{"amount":0.000005,"unit":"outputToken","label":"≤ 272K context"},{"amount":0.0000075,"unit":"outputToken","label":"> 272K context"},{"amount":5e-8,"unit":"cachedInputToken","label":"≤ 272K context"},{"amount":1e-7,"unit":"cachedInputToken","label":"> 272K context"},{"amount":0.00000125,"unit":"cacheWriteToken","label":"≤ 272K context"},{"amount":0.0000025,"unit":"cacheWriteToken","label":"> 272K context"}]},{"model":"klingai-video-4-0","air":"klingai:kling-video@4.0","name":"Kling VIDEO 4.0","creator":"klingai","status":"live","releasedAt":"2026-10-23T00:00:00Z","weight":5,"headline":"Unified video generation and editing with multi-keyframe control and clips up to 30 seconds","description":"Kling VIDEO 4.0 is a unified video model that generates 1080p clips of up to 30 seconds from text, images, reference videos, and reusable elements. It follows up to 10 keyframes per clip, including first, middle, and last frames, and combines reference images, reference videos, and elements in a single request to hold subjects, style, and motion. It also edits an existing video from a prompt, which suits long narrative shots, storyboard-driven sequences, and reference-driven production work.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-4-0.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-4-0.png","pricingRates":[{"amount":0.157,"unit":"durationSecond","label":"720p · No video input"},{"amount":0.235,"unit":"durationSecond","label":"1080p · No video input"},{"amount":0.188,"unit":"durationSecond","label":"720p · Video input"},{"amount":0.282,"unit":"durationSecond","label":"1080p · Video input"}],"pricingExamples":[{"configuration":"video-edit","price":"$1.70","amount":1.7049,"latencyMs":140209,"exampleId":"adaptive-prosthetic-fashion-showcase"},{"configuration":"first-frame","price":"$2.35","amount":2.35,"latencyMs":1655405,"exampleId":"technical-apiary-suit-product-reel"},{"configuration":"elements","price":"$2.35","amount":2.35,"latencyMs":164764,"exampleId":"seed-vault-film-stock-portrait"},{"configuration":"first-last-frame","price":"$2.35","amount":2.35,"latencyMs":476551,"exampleId":"urban-tree-pathology-presenter-clip"},{"configuration":"text-to-video","price":"$2.35","amount":2.35,"latencyMs":616355,"exampleId":"whale-skeleton-rig-cleanup-plate"},{"configuration":"multiple-keyframes","price":"$2.35","amount":2.35,"latencyMs":212322,"exampleId":"subsea-monitoring-logo-animation"},{"configuration":"reference-to-video","price":"$2.82","amount":2.82,"latencyMs":192369,"exampleId":"glacial-handpan-performance-visualizer"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Kling VIDEO 4.0, from one-line ideas to structured briefs with timed beats, camera direction, and an audio section that sets what you hear.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/prompting"},{"slug":"dialogue-and-on-screen-text","title":"Dialogue and on-screen text","description":"How to write spoken lines, languages and accents for Kling VIDEO 4.0, and how to get readable signs and burned-in subtitles that hold steady while the camera moves.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/dialogue-and-on-screen-text"},{"slug":"long-takes-and-multi-shot","title":"Long takes and multi-shot sequences","description":"How to use Kling VIDEO 4.0's 30-second range for complete stories, either as a timed shot list with hard cuts or as one continuous take with no cuts at all.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/long-takes-and-multi-shot"},{"slug":"keyframes","title":"Keyframes","description":"How to anchor Kling VIDEO 4.0 to still images: animate a first frame, bridge a first and last frame, place up to ten keyframes on a timeline, and choose the output shape.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/keyframes"},{"slug":"reference-images","title":"Reference images","description":"How to drive Kling VIDEO 4.0 with reference images: give each image a role, combine people with products and locations, keep a cast consistent, and follow a storyboard.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/reference-images"},{"slug":"reference-videos","title":"Reference videos","description":"How to use reference videos with Kling VIDEO 4.0 to carry a routine, a camera path, or the structure of an ad over to new people, products, and locations.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/reference-videos"},{"slug":"editing-video","title":"Editing video","description":"How to edit existing footage with Kling VIDEO 4.0: swap a product for a reference image, add or remove a subject, or change the visual style while the shot stays the same.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/editing-video"},{"slug":"elements","title":"Elements","description":"How to save a character in Kling VIDEO 4.0 as an element under your own id, build it from several angles or a short clip, and reuse it across a series of videos.","date":"2026-10-22","url":"https://runware.ai/docs/models/klingai-video-4-0/guides/elements"}]},{"model":"heygen-voice-1-0","air":"heygen:voice@1.0","name":"HeyGen Voice 1.0","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-10-22T00:00:00Z","weight":5,"headline":"Text-to-speech with public voices and instant voice cloning from a single recording","description":"HeyGen Voice 1.0 is HeyGen's text-to-speech model for spoken audio in 39 languages, with the language detected from the text when none is set. It speaks with a public voice or clones a new one from a single recording, and returns the cloned voice so later requests can reuse it. It suits voiceovers, narration for training and product videos, and localized content.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/heygen-voice-1-0.jpg","ogImage":"https://assets.runware.ai/og/heygen-voice-1-0.png","pricingRates":[{"amount":0.00003,"unit":"character"}],"pricingExamples":[{"configuration":"voice-cloning","price":"$0.00304","amount":0.003045,"latencyMs":13652,"exampleId":"dutch-flower-auction-lot-call"},{"configuration":"text-to-audio","price":"$0.00715","amount":0.007155,"latencyMs":12674,"exampleId":"polar-seed-vault-pa-announcement"},{"configuration":"voice-cloning","price":"$0.00792","amount":0.00792,"latencyMs":22313,"exampleId":"perfume-archive-handling-course-narration"},{"configuration":"text-to-audio","price":"$0.0108","amount":0.010755,"latencyMs":7423,"exampleId":"ballast-water-compliance-explainer-voiceover"}],"guides":[{"slug":"public-voices","title":"Choosing a public voice","description":"How to pick one of HeyGen Voice 1.0's 24 public voices for a script, from ad reads and training narration to support lines, wellness audio and announcements.","date":"2026-10-23","url":"https://runware.ai/docs/models/heygen-voice-1-0/guides/public-voices"},{"slug":"voice-cloning","title":"Cloning a voice from a recording","description":"How to clone a speaker with HeyGen Voice 1.0 from one recording, reuse the voice in later requests through outputs.voiceId, and set expressiveness for a closer match.","date":"2026-10-23","url":"https://runware.ai/docs/models/heygen-voice-1-0/guides/voice-cloning"},{"slug":"languages","title":"Speaking other languages","description":"How to make HeyGen Voice 1.0 speak 39 languages with a public or cloned voice, when to set speech.language, and how to localize with one voice across markets.","date":"2026-10-23","url":"https://runware.ai/docs/models/heygen-voice-1-0/guides/languages"}]},{"model":"heygen-voice-1-0-turbo","air":"heygen:voice@1.0-turbo","name":"HeyGen Voice 1.0 Turbo","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-10-22T00:00:00Z","weight":5,"headline":"Faster HeyGen text-to-speech with public voices and instant voice cloning","description":"HeyGen Voice 1.0 Turbo is the faster variant of HeyGen Voice 1.0, generating speech in 39 languages from text. It uses the same public voices and the same instant cloning from a single recording, and returns the cloned voice so later requests can reuse it. It suits high-volume narration and pipelines where turnaround matters.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/heygen-voice-1-0-turbo.jpg","ogImage":"https://assets.runware.ai/og/heygen-voice-1-0-turbo.png","pricingRates":[{"amount":0.000004,"unit":"character"}],"pricingExamples":[{"configuration":"voice-cloning","price":"$0.000368","amount":0.000368,"latencyMs":23926,"exampleId":"tool-library-vocal-brand-stinger"},{"configuration":"text-to-audio","price":"$0.000388","amount":0.000388,"latencyMs":3388,"exampleId":"roller-derby-broadcast-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.000812","amount":0.000812,"latencyMs":5123,"exampleId":"subterranean-seed-vault-staff-announcement"},{"configuration":"voice-cloning","price":"$0.00334","amount":0.003344,"latencyMs":17665,"exampleId":"historic-neon-conservation-training-narration"}]},{"model":"heygen-voice-1-0-turbo-free","air":"heygen:voice@1.0-turbo-free","name":"HeyGen Voice 1.0 Turbo Free","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-10-22T00:00:00Z","weight":5,"headline":"HeyGen Voice Turbo on shared capacity for tests and prototypes","description":"HeyGen Voice 1.0 Turbo Free runs HeyGen Voice 1.0 Turbo on shared, limited capacity. It offers the same public voices, instant voice cloning from a single recording and 39 languages, but response time and availability are not guaranteed. It suits tests, prototypes and small amounts of speech.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/heygen-voice-1-0-turbo-free.jpg","ogImage":"https://assets.runware.ai/og/heygen-voice-1-0-turbo-free.png"},{"model":"ideogram-vector-1-0","air":"ideogram:vector@1.0","name":"Ideogram Vector 1.0","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-10-22T00:00:00Z","weight":5,"headline":"Scalable, editable SVG artwork from text prompts or existing images","description":"Ideogram Vector 1.0 is Ideogram's vector model, generating SVG artwork from text prompts, optionally guided by up to five reference images, and converting existing images into SVG. Each shape stays editable, so colors can be matched to a brand and individual elements refined in a vector editor. It suits logos, typography, illustrations, print designs, and labels or stickers.","capabilities":["io:text-to-image","io:image-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-vector-1-0.jpg","ogImage":"https://assets.runware.ai/og/ideogram-vector-1-0.png","pricingRates":[{"amount":0.13,"unit":"output","label":"Text to vector · Low"},{"amount":0.16,"unit":"output","label":"Text to vector · Medium"},{"amount":0.2,"unit":"output","label":"Text to vector · High"},{"amount":0.32,"unit":"output","label":"Text to vector · High · Reference images"},{"amount":0.1,"unit":"output","label":"Image to vector"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.10","amount":0.1,"latencyMs":15555,"exampleId":"steam-fire-engine-coloring-plate"},{"configuration":"vectorize","price":"$0.10","amount":0.1,"latencyMs":9204,"exampleId":"vaccine-carrier-field-lifestyle-vector"},{"configuration":"reference-to-vector · 1536×2048","price":"$0.16","amount":0.16,"latencyMs":41245,"exampleId":"cave-robotics-logo-vectorization"},{"configuration":"text-to-vector · 2048×2048 · quality high","price":"$0.20","amount":0.2,"latencyMs":71679,"exampleId":"water-smart-yard-rebate-campaign"},{"configuration":"text-to-vector · 2048×2048 · quality high","price":"$0.20","amount":0.2,"latencyMs":45147,"exampleId":"mycelium-acoustics-brand-logo"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Ideogram Vector 1.0: describing shapes and a palette, quoting exact lettering, and picking a size and a quality tier.","date":"2026-10-29","url":"https://runware.ai/docs/models/ideogram-vector-1-0/guides/prompting"},{"slug":"vectorizing-images","title":"Vectorizing images","description":"How to convert raster images to SVG with Ideogram Vector 1.0: the request, the size of the result, and the simple, detailed and auto modes.","date":"2026-10-29","url":"https://runware.ai/docs/models/ideogram-vector-1-0/guides/vectorizing-images"},{"slug":"reference-images","title":"Reference images","description":"How to guide Ideogram Vector 1.0 with reference images: redrawing a sketch or a photo as vector art, giving each image a role, and setting the shape.","date":"2026-10-29","url":"https://runware.ai/docs/models/ideogram-vector-1-0/guides/reference-images"}]},{"model":"prunaai-p-video-2-pro-avatar","air":"prunaai:p-video@2-pro-avatar","name":"P-Video-2-Pro-Avatar","creator":"prunaai","status":"live","releasedAt":"2026-10-22T00:00:00Z","weight":5,"headline":"Lip-synced talking video from one portrait and an audio track or script, up to 3 minutes and 4K","description":"P-Video-2-Pro-Avatar is Pruna AI's talking-video model that turns a single portrait into a lip-synced video driven by an uploaded audio track or a script read by built-in text-to-speech. Lips and expression follow the soundtrack, the video runs as long as the speech, from 1 second to 3 minutes, and a prompt can steer motion and expression. It offers 30 voices across 10 languages with delivery instructions for tone and pacing, and outputs from 480p up to 4K for course instructors, localized product walkthroughs, and personalized outreach.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-2-pro-avatar.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-2-pro-avatar.png","pricingRates":[{"amount":0.02,"unit":"durationSecond","label":"480p"},{"amount":0.04,"unit":"durationSecond","label":"768p"},{"amount":0.06,"unit":"durationSecond","label":"1080p"},{"amount":0.08,"unit":"durationSecond","label":"2K"},{"amount":0.12,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"audio-driven-photo-avatar · 1080p","price":"$0.241","amount":0.2413,"latencyMs":75535,"exampleId":"wetland-technician-documentary-soundbite"},{"configuration":"audio-driven-photo-avatar · 1080p","price":"$0.539","amount":0.5387,"latencyMs":129292,"exampleId":"seed-vault-sensor-explainer"},{"configuration":"scripted-photo-avatar · 4k","price":"$1.05","amount":1.055,"latencyMs":273489,"exampleId":"gemstone-provenance-intake-briefing"},{"configuration":"scripted-photo-avatar · 4k","price":"$1.29","amount":1.29,"latencyMs":302729,"exampleId":"nocturnal-flower-market-lyric-interlude"}]},{"model":"bria-fibo-2-0","air":"bria:20@1","name":"Bria FIBO 2.0","creator":"bria","status":"live","releasedAt":"2026-10-20T00:00:00Z","weight":5,"headline":"Structured-prompt image generation from 5:1 to 1:5, up to 4MP","description":"Bria FIBO 2.0 is a text-to-image and image-to-image model that turns a prompt or a reference image into a structured JSON prompt before rendering, which makes results reproducible and easy to refine. It renders 21 aspect ratios from 5:1 to 1:5 at 1MP, and the nine standard ratios at 4MP. It is trained only on licensed data, which suits commercial production work.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-fibo-2-0.jpg","ogImage":"https://assets.runware.ai/og/bria-fibo-2-0.png","pricingOverview":"Each image costs $0.04 at any supported size, up to 4MP.","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1568×672","price":"$0.04","amount":0.04,"latencyMs":21177,"exampleId":"artistic-swim-team-press-portrait"},{"configuration":"image-edit · 1536×2048","price":"$0.04","amount":0.04,"latencyMs":56929,"exampleId":"seed-vault-manga-background-plate"},{"configuration":"text-to-image · 1568×672","price":"$0.04","amount":0.04,"latencyMs":18989,"exampleId":"mammoth-exhibition-atrium-composite"},{"configuration":"image-edit · 1536×2048","price":"$0.04","amount":0.04,"latencyMs":49198,"exampleId":"countertop-composter-campaign-visual"}]},{"model":"bria-fibo-2-0-edit","air":"bria:21@1","name":"Bria FIBO 2.0 Edit","creator":"bria","status":"live","releasedAt":"2026-10-20T00:00:00Z","weight":5,"headline":"Instruction-based editing with masks and up to four reference images","description":"Bria FIBO 2.0 Edit is an image editing model that changes an image from a text instruction, with an optional mask to regenerate one area and keep the rest. It also combines up to four reference images into one result at a chosen aspect ratio, for product placement, color or texture matching, and character consistency. It is trained only on licensed data, which suits commercial production work.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-fibo-2-0-edit.jpg","ogImage":"https://assets.runware.ai/og/bria-fibo-2-0-edit.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 1408×792","price":"$0.04","amount":0.04,"latencyMs":19496,"exampleId":"opera-conductor-headshot-display"},{"configuration":"inpaint","price":"$0.04","amount":0.04,"latencyMs":14937,"exampleId":"apiary-smoker-lifestyle-product-photo"},{"configuration":"image-edit","price":"$0.04","amount":0.04,"latencyMs":15320,"exampleId":"polar-research-annual-report-hero"},{"configuration":"reference-to-image · 1408×792","price":"$0.04","amount":0.04,"latencyMs":19784,"exampleId":"boutique-pedalboard-product-group-shot"}]},{"model":"minimax-m3-1-flash","air":"minimax:m3.1@flash","name":"MiniMax M3.1 Flash","creator":"minimax","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-10-20T00:00:00Z","weight":5,"headline":"Efficient open-weight multimodal model for coding, agents, and 1M-token context","description":"MiniMax M3.1 Flash is an efficient open-weight model in MiniMax's M-series for everyday coding, agentic reasoning, tool use, and long-context work. It accepts text, images, and video, with a 1M-token context window, up to 524,288 output tokens, and reasoning depth that can be tuned but not turned off. It suits coding agents, multi-step workflow automation, chart and document analysis, and frontend or presentation generation.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-m3-1-flash.jpg","ogImage":"https://assets.runware.ai/og/minimax-m3-1-flash.png","contextWindow":1000000,"maxOutputTokens":524288,"pricingBasis":"token","pricingRates":[{"amount":1.5e-7,"unit":"inputToken"},{"amount":6e-7,"unit":"outputToken"},{"amount":1.5e-8,"unit":"cachedInputToken"}]},{"model":"deepseek-v4-1-flash-1017","air":"deepseek:v4.1@flash-1017","name":"DeepSeek-V4.1-Flash-1017","creator":"deepseek","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-10-17T00:00:00Z","weight":5,"headline":"Multimodal 1M-context MoE checkpoint for coding, reasoning, and tool-using agents","description":"DeepSeek-V4.1-Flash-1017 is the October 2026 checkpoint of DeepSeek-V4.1-Flash, a multimodal mixture-of-experts model for coding, reasoning, visual understanding, and tool-using agents. It reads text and images, writes text, and combines a 1M-token context window and up to 1M output tokens with continuously adjustable reasoning effort and tool calls. It suits large-codebase work, document analysis, research, and long-running agentic tasks.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/deepseek-v4-1-flash-1017.jpg","ogImage":"https://assets.runware.ai/og/deepseek-v4-1-flash-1017.png","contextWindow":1048576,"maxOutputTokens":384000,"pricingBasis":"token","pricingRates":[{"amount":1.5e-7,"unit":"inputToken"},{"amount":6e-7,"unit":"outputToken"},{"amount":1e-8,"unit":"cachedInputToken"}]},{"model":"bfl-flux-3-image-klein","air":"bfl:flux@3-image-klein","name":"FLUX 3 Image Klein","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-10-15T00:00:00Z","weight":5,"headline":"Distilled open-weights model for fast image generation and editing","description":"FLUX 3 Image Klein is the distilled checkpoint of Black Forest Labs' open-weights FLUX 3 image model, covering text-to-image generation and reference-based editing in a single model. It is built for speed without giving up quality, which suits interactive applications, rapid iteration, and high-volume production.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-3-image-klein.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-3-image-klein.png","pricingMeasured":[{"configuration":"512×512 · steps 16","price":0.003,"latencyMs":2528},{"configuration":"1024×1024 · steps 16","price":0.006,"latencyMs":2869},{"configuration":"2048×2048 · steps 16","price":0.0179,"latencyMs":7440},{"configuration":"4096×4096 · steps 16","price":0.1013,"latencyMs":36188}],"pricingExamples":[{"configuration":"text-to-image · 1344×768","price":"$0.006","amount":0.006,"latencyMs":3335,"exampleId":"arctic-research-director-headshot"},{"configuration":"bounding-boxes · 1536×1024","price":"$0.0089","amount":0.0089,"latencyMs":4012,"exampleId":"desert-railcar-resort-composite"},{"configuration":"reference-bounding-boxes · 1024×1536","price":"$0.0149","amount":0.0149,"latencyMs":8918,"exampleId":"adaptive-tailoring-lookbook-portrait"},{"configuration":"reference-to-image · 768×1344","price":"$0.0149","amount":0.0149,"latencyMs":9931,"exampleId":"apiary-suit-virtual-try-on"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt FLUX 3 Image Klein for text-to-image: writing a full scene caption, holding a layout with the seed, sizing up to 4K, and trading steps for speed.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein/guides/prompting"},{"slug":"text-rendering","title":"Rendering text","description":"How to render readable text with FLUX 3 Image Klein: quoting the exact words, controlling the text you didn't ask for, building a layout hierarchy, and editing signage.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein/guides/text-rendering"},{"slug":"editing-images","title":"Editing images","description":"How to edit a photo with FLUX 3 Image Klein: scoping the change, removing objects, writing scene-wide edits as captions, reframing, and chaining edits.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein/guides/editing-images"},{"slug":"multi-reference-composition","title":"Multi-reference composition","description":"How to combine up to four references with FLUX 3 Image Klein: assigning roles, keeping a product across scenes, placing it in your own photo, and borrowing a style.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein/guides/multi-reference-composition"},{"slug":"bounding-boxes","title":"Bounding boxes","description":"How to place, move and remove elements by region with FLUX 3 Image Klein: the 0 to 1000 grid, laying out a frame, reserving space for copy, and editing by box.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein/guides/bounding-boxes"}]},{"model":"bfl-flux-3-image-klein-base","air":"bfl:flux@3-image-klein-base","name":"FLUX 3 Image Klein Base","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-10-15T00:00:00Z","weight":5,"headline":"Undistilled open-weights foundation for fine-tuning and customization","description":"FLUX 3 Image Klein Base is the full undistilled checkpoint of Black Forest Labs' open-weights FLUX 3 image model, covering text-to-image generation and reference-based editing. It keeps the complete model capacity, which makes it the variant to fine-tune, customize, and optimize. It suits research, post-training work, and workflows that favor control over generation speed.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-3-image-klein-base.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-3-image-klein-base.png","pricingMeasured":[{"configuration":"512×512 · steps 50","price":0.0089,"latencyMs":3989},{"configuration":"1024×1024 · steps 50","price":0.0238,"latencyMs":8504},{"configuration":"2048×2048 · steps 50","price":0.0924,"latencyMs":33423},{"configuration":"4096×4096 · steps 50","price":0.5334,"latencyMs":182208}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.0238","amount":0.0238,"latencyMs":8846,"exampleId":"oyster-farm-spectrometer-lifestyle-photo"},{"configuration":"bounding-boxes · 1280×768","price":"$0.0238","amount":0.0238,"latencyMs":9029,"exampleId":"subsea-cable-recruitment-key-art"},{"configuration":"reference-bounding-boxes · 1024×1280","price":"$0.0536","amount":0.0536,"latencyMs":21042,"exampleId":"dark-sky-retrofit-social-media-post"},{"configuration":"reference-to-image · 1024×1280","price":"$0.0685","amount":0.0685,"latencyMs":28480,"exampleId":"volcanic-observatory-director-portrait"}],"guides":[{"slug":"guidance-and-negative-prompts","title":"Guidance and negative prompts","description":"How to use CFG scale and negative prompts with FLUX 3 Image Klein Base: how guidance changes adherence, what a negative prompt removes, and what step count to run.","date":"2026-10-20","url":"https://runware.ai/docs/models/bfl-flux-3-image-klein-base/guides/guidance-and-negative-prompts"}]},{"model":"krea-3","air":"krea:krea@3","name":"Krea 3","creator":"krea","status":"live","releasedAt":"2026-10-15T00:00:00Z","weight":5,"headline":"Krea image model for precise region edits, reference-guided generation, and up to 4K output","description":"Krea 3 is Krea's image model for text-to-image generation and precise editing of existing images. Boxes mark the areas to change and the objects to bring in from up to eight reference images, each with an optional label and instruction, while the rest of the frame stays untouched. Output comes in 1K, 2K, and 4K across ten aspect ratios, with a fast or high thinking level.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-3.jpg","ogImage":"https://assets.runware.ai/og/krea-3.png","pricingRates":[{"amount":0.025,"unit":"outputMegapixel","label":"Output"},{"amount":0.025,"unit":"outputMegapixel","label":"Each reference image"}],"pricingExamples":[{"configuration":"text-to-image · 1360×768","price":"$0.0125","amount":0.012451,"latencyMs":10846,"exampleId":"canal-house-hotel-website-hero"},{"configuration":"reference-bounding-boxes · 2K","price":"$0.15","amount":0.15,"latencyMs":54159,"exampleId":"antique-auction-banner-expansion"},{"configuration":"reference-to-image · 2K","price":"$0.15","amount":0.15,"latencyMs":40544,"exampleId":"peatland-science-mug-catalog-photograph"},{"configuration":"local-editing · 4K","price":"$0.40","amount":0.4,"latencyMs":48666,"exampleId":"bicycle-chain-wax-tin-retouch"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Krea 3: layering a scene description, choosing one of the 30 output sizes and deciding when the high thinking level is worth the wait.","date":"2026-11-04","url":"https://runware.ai/docs/models/krea-3/guides/prompting"},{"slug":"editing-images","title":"Editing images","description":"How to edit a photo with Krea 3 using an instruction alone: relighting, replacing a background, changing the season, restyling, and reframing to a new aspect ratio.","date":"2026-11-04","url":"https://runware.ai/docs/models/krea-3/guides/editing-images"},{"slug":"reference-images","title":"Reference images","description":"How to compose new images from up to eight references with Krea 3: naming each image, placing products in scenes, keeping a person consistent and borrowing a style.","date":"2026-11-04","url":"https://runware.ai/docs/models/krea-3/guides/reference-images"},{"slug":"edit-regions","title":"Edit regions","description":"How to edit parts of an image with Krea 3 using settings.editRegions: box coordinates, labels and notes, several regions at once, objects from another reference, and printed text.","date":"2026-11-04","url":"https://runware.ai/docs/models/krea-3/guides/edit-regions"}]},{"model":"typesafe-jev-1-14","air":"typesafe:jev@1.14","name":"Jev 1.14","creator":"typesafe-ai","endpoints":["decisions"],"status":"compatible-only","releasedAt":"2026-10-14T00:00:00Z","weight":5,"headline":"System One decision model for typed answers over long, multilingual application state","description":"Jev 1.14 is TypeSafe AI's System One decision model for fast, machine-consumable judgments. It reads structured or unstructured application state and answers typed questions with bounded choices, ordinal scores, or yes-or-no probabilities with confidence, with stronger multilingual handling. Its 256K-token context window fits long logs, documents, and conversation histories into a single decision for routing, classification, verification, and guardrail workflows.","capabilities":["io:text-to-text","op:classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/typesafe-jev-1-14.jpg","ogImage":"https://assets.runware.ai/og/typesafe-jev-1-14.png","contextWindow":262144,"pricingBasis":"token","pricingRates":[{"amount":4.2e-8,"unit":"inputToken"}]},{"model":"google-gemini-4-argon","air":"google:gemini@4-argon","name":"Gemini 4 Argon","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-10-13T00:00:00Z","weight":5,"headline":"Google's frontier Gemini model for long-horizon coding, enterprise knowledge work, and cyber defense","description":"Gemini 4 Argon is Google's frontier model for complex, long-horizon work across software engineering, enterprise knowledge work, and cybersecurity defense. It accepts text, images, video, audio, and PDFs, with a 1M-token context window, up to 32,768 output tokens, and configurable thinking effort. It is well suited to large codebase refactors and migrations, document-heavy legal and financial analysis, long-video understanding, and agent workflows that need deep reasoning and precise multi-step execution.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-4-argon.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-4-argon.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"Through Nov 30, 2026"},{"amount":0.000004,"unit":"inputToken","label":"From Dec 1, 2026"},{"amount":0.00001,"unit":"outputToken","label":"Through Nov 30, 2026"},{"amount":0.00002,"unit":"outputToken","label":"From Dec 1, 2026"},{"amount":1e-7,"unit":"cachedInputToken","label":"Through Nov 30, 2026"},{"amount":2e-7,"unit":"cachedInputToken","label":"From Dec 1, 2026"}]},{"model":"alibaba-qwen-image-2-1-turbo","air":"alibaba:qwen-image@2.1-turbo","name":"Qwen-Image-2.1-Turbo","creator":"alibaba","status":"live","releasedAt":"2026-10-09T00:00:00Z","weight":5,"headline":"Low-cost image generation and multi-reference editing with LLM prompt rewriting","description":"Qwen-Image-2.1-Turbo is the lower-cost Turbo tier of Alibaba's Qwen-Image-2.1 family, a unified model for text-to-image generation and prompt-guided image editing. It composes and edits from up to 10 reference images and can rewrite prompts with an LLM before generation, in a direct or agent mode with optional thinking. It outputs from 0.26 to 4.19 megapixels at aspect ratios from 1:8 to 8:1, suited to high-volume product imagery, marketing visuals, and rapid iteration.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-2-1-turbo.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-2-1-turbo.png","pricingRates":[{"amount":0.014133,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 1536×1024","price":"$0.0141","amount":0.014133,"latencyMs":23645,"exampleId":"wayang-exhibition-campaign-hero"},{"configuration":"text-to-image · 1024×1024","price":"$0.0141","amount":0.014133,"latencyMs":17275,"exampleId":"vintage-rescue-helicopter-arctic-composite"},{"configuration":"reference-to-image · 1536×1024","price":"$0.0141","amount":0.014133,"latencyMs":30924,"exampleId":"forensic-entomologist-creator-avatar"},{"configuration":"text-to-image · 1024×1024","price":"$0.0141","amount":0.014133,"latencyMs":23190,"exampleId":"geothermal-district-heating-explainer"}]},{"model":"anthropic-claude-haiku-5-5","air":"anthropic:claude@haiku-5.5","name":"Claude Haiku 5.5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-10-07T00:00:00Z","weight":5,"headline":"Fastest, lowest-cost Claude model for high-volume agents, extraction, and low-latency applications","description":"Claude Haiku 5.5 is the fastest and lowest-cost model in Anthropic’s Claude 5.5 family, built for high-volume and latency-sensitive work such as classification, extraction, routing, and customer support. It reads text and images, calls tools, and offers configurable effort, so a request can trade response time against depth. Its 1M-token context window and 128K-token maximum output also make it a practical sub-agent inside larger coding and research workflows.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-haiku-5-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-haiku-5-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":1e-7,"unit":"inputToken","label":"≤ 100K context"},{"amount":5e-7,"unit":"inputToken","label":"> 100K context"},{"amount":5e-7,"unit":"outputToken","label":"≤ 100K context"},{"amount":0.0000025,"unit":"outputToken","label":"> 100K context"},{"amount":1e-8,"unit":"cachedInputToken","label":"≤ 100K context"},{"amount":5e-8,"unit":"cachedInputToken","label":"> 100K context"},{"amount":1.25e-7,"unit":"cacheWriteToken","label":"≤ 100K context"},{"amount":6.25e-7,"unit":"cacheWriteToken","label":"> 100K context"}]},{"model":"google-nano-banana-2-1","air":"google:nano-banana@2.1","name":"Nano Banana 2.1","creator":"google","zdr":true,"status":"live","releasedAt":"2026-10-06T00:00:00Z","weight":5,"headline":"Sharper graphic composition and subject consistency, with multi-reference input and live search grounding","description":"Nano Banana 2.1 is Google's updated image generation and editing model in the Nano Banana 2 family. It improves graphic composition and subject consistency, and it follows complex prompts and edit instructions more accurately. It works from text alone or from as many as fourteen reference images plus one reference video, and it can ground a generation in live web and image search so the result reflects current facts and real visual references. It generates at 1K, 2K and 4K, with three levels of thinking that trade speed for reasoning depth, which suits layout-heavy design work, recurring characters and products, and precise multi-step edits.","capabilities":["io:text-to-image","io:image-to-image","io:video-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-nano-banana-2-1.jpg","ogImage":"https://assets.runware.ai/og/google-nano-banana-2-1.png","pricingBasis":"floor","pricingRates":[{"amount":0.0000015,"unit":"inputToken"},{"amount":0.0000075,"unit":"outputToken","label":"Text output"},{"amount":0.00003,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"text-to-image · 1376×768","price":"$0.0341","amount":0.034098,"latencyMs":13098,"exampleId":"autumn-dusk-memorial-garden-visualization"},{"configuration":"text-to-image · 1376×768","price":"$0.0342","amount":0.0342165,"latencyMs":32103,"exampleId":"geothermal-plant-operations-icon-set"},{"configuration":"video-reference · 1376×768","price":"$0.0346","amount":0.0345765,"latencyMs":49530,"exampleId":"pipe-organ-restoration-social-still"},{"configuration":"image-to-image · 2K","price":"$0.0524","amount":0.0523815,"latencyMs":24650,"exampleId":"marine-radio-housing-colorway-variant"},{"configuration":"identity-preservation · 2K","price":"$0.0544","amount":0.0543795,"latencyMs":21751,"exampleId":"tidal-sound-festival-press-portrait"},{"configuration":"search-grounded · 1200×896","price":"$0.0905","amount":0.0905495,"latencyMs":35054,"exampleId":"artemis-ii-crew-shirt-graphic"},{"configuration":"reference-to-image · 4K","price":"$0.122","amount":0.121659,"latencyMs":85269,"exampleId":"antarctic-snowcat-exhibition-key-visual"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Nano Banana 2.1: briefs with counts and positions, layered descriptions, a system prompt for a house style, thinking levels, and sizes.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/prompting"},{"slug":"text-and-layouts","title":"Text and layouts","description":"How to get exact copy and structured layouts from Nano Banana 2.1: quoting strings, setting a type hierarchy, building infographics, and translating a design.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/text-and-layouts"},{"slug":"editing-images","title":"Editing images","description":"How to edit an image with Nano Banana 2.1: one reference and a written instruction, keep lists, removals and additions, numbered changes, and chained edits.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/editing-images"},{"slug":"multi-reference-composition","title":"Composing from multiple reference images","description":"How to combine up to 14 reference images with Nano Banana 2.1: giving each image a role, following a layout sketch, borrowing a style, and reading a video.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/multi-reference-composition"},{"slug":"subject-consistency","title":"Subject consistency","description":"How to keep one product or character identical across a series with Nano Banana 2.1: one reference into new scenes, extra views, and illustrated characters.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/subject-consistency"},{"slug":"grounded-generation","title":"Grounding with web and image search","description":"How to ground Nano Banana 2.1 in live facts and real appearances with settings.webSearch and settings.imageSearch, and when to leave both lookups off.","date":"2026-10-06","url":"https://runware.ai/docs/models/google-nano-banana-2-1/guides/grounded-generation"}]},{"model":"bfl-flux-3-image","air":"bfl:flux@3-image","name":"FLUX 3 Image","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-10-01T00:00:00Z","weight":5,"headline":"Native 4K image generation and editing with precise multi-reference control","description":"FLUX 3 Image is Black Forest Labs' image generation and editing model built on the multimodal FLUX 3 backbone shared with FLUX 3 Video. It combines text-to-image synthesis, precise local editing, multi-reference composition, bounding-box placement, and native 4K output. The model preserves identities and fine details across references, supports targeted changes and in-place text editing, and renders accurate typography in text-heavy layouts across a broad range of visual styles.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-3-image.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-3-image.png","pricingRates":[{"amount":0.041,"unit":"output","label":"0.75K"},{"amount":0.048,"unit":"output","label":"1K"},{"amount":0.1,"unit":"output","label":"2K"},{"amount":0.607,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"bounding-boxes · 1360×768","price":"$0.024","amount":0.024,"latencyMs":50189,"exampleId":"circular-linen-campaign-key-visual"},{"configuration":"text-to-image · 1360×768","price":"$0.048","amount":0.048,"latencyMs":44132,"exampleId":"rare-manuscript-conservator-portrait"},{"configuration":"text-to-image · 1360×768","price":"$0.048","amount":0.048,"latencyMs":40011,"exampleId":"geothermal-utility-campaign-key-visual"},{"configuration":"identity-preservation · 2K","price":"$0.05","amount":0.05,"latencyMs":46423,"exampleId":"grain-inspector-staff-profile-portrait"},{"configuration":"background-replacement · 2K","price":"$0.05","amount":0.05,"latencyMs":80172,"exampleId":"maritime-museum-robot-headshot"},{"configuration":"reference-bounding-boxes · 2K","price":"$0.05","amount":0.05,"latencyMs":59230,"exampleId":"library-returns-station-catalogue"},{"configuration":"reference-to-image · 2K","price":"$0.10","amount":0.1,"latencyMs":363076,"exampleId":"seed-library-supper-invitation"},{"configuration":"reference-to-image · 2K","price":"$0.10","amount":0.1,"latencyMs":295841,"exampleId":"floating-sauna-social-campaign-post"},{"configuration":"local-editing · 4K","price":"$0.303","amount":0.3035,"latencyMs":99885,"exampleId":"festival-poster-title-retouch"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt FLUX 3 Image for text-to-image: the five-part prompt structure, the detail that prompt expansion cannot invent for you, camera direction, and output size.","date":"2026-10-01","url":"https://runware.ai/docs/models/bfl-flux-3-image/guides/prompting"},{"slug":"text-rendering","title":"Rendering text","description":"How to render readable text with FLUX 3 Image: quoting the exact string, directing the typography, building a text hierarchy, and editing copy inside an existing image.","date":"2026-10-01","url":"https://runware.ai/docs/models/bfl-flux-3-image/guides/text-rendering"},{"slug":"editing-images","title":"Editing images","description":"How to edit an image with FLUX 3 Image from a single reference: scoping the instruction, swapping materials and light, removing objects, reframing, and chaining edits.","date":"2026-10-01","url":"https://runware.ai/docs/models/bfl-flux-3-image/guides/editing-images"},{"slug":"multi-reference-composition","title":"Multi-reference composition","description":"How to combine up to ten reference images with FLUX 3 Image: assigning a role to each one, holding a face or a product across scenes, and borrowing a look.","date":"2026-10-01","url":"https://runware.ai/docs/models/bfl-flux-3-image/guides/multi-reference-composition"},{"slug":"bounding-boxes","title":"Bounding boxes","description":"How to place and edit elements by region with FLUX 3 Image: the 0 to 1000 grid, composing a layout from scratch, moving a marked element, and removing one.","date":"2026-10-01","url":"https://runware.ai/docs/models/bfl-flux-3-image/guides/bounding-boxes"}]},{"model":"heygen-video-1-0","air":"heygen:video@1.0","name":"HeyGen Video 1.0","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-09-30T00:00:00Z","weight":5,"headline":"Prompt-driven video generation with native dialogue, ambience, and sound effects","description":"HeyGen Video 1.0 is HeyGen's video generation model for short clips with synchronized audio. It generates the full frame from a text prompt, a first-frame image, or up to 12 image, video, and audio references, and produces dialogue, ambience, and sound effects in the same pass, with no separate lip-sync step. It outputs 5 to 30 second clips at 480p, 768p, 1080p, or 2K, suited to presenter shots, product and training content, and social video.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/heygen-video-1-0.jpg","ogImage":"https://assets.runware.ai/og/heygen-video-1-0.png","pricingRates":[{"amount":0.02,"unit":"durationSecond","label":"480p, from text or an image"},{"amount":0.036,"unit":"durationSecond","label":"768p, from text or an image"},{"amount":0.1,"unit":"durationSecond","label":"1080p, from text or an image"},{"amount":0.1,"unit":"durationSecond","label":"2K, from text or an image"},{"amount":0.07,"unit":"durationSecond","label":"480p, from references"},{"amount":0.1,"unit":"durationSecond","label":"768p, from references"},{"amount":0.18,"unit":"durationSecond","label":"1080p, from references"},{"amount":0.18,"unit":"durationSecond","label":"2K, from references"},{"amount":0.07,"unit":"durationSecond","label":"480p, per second of reference video, added to the output"},{"amount":0.1,"unit":"durationSecond","label":"768p, per second of reference video, added to the output"},{"amount":0.18,"unit":"durationSecond","label":"1080p, per second of reference video, added to the output"},{"amount":0.18,"unit":"durationSecond","label":"2K, per second of reference video, added to the output"},{"amount":0.04,"unit":"inputImage","label":"Each reference image after the first two"}],"pricingExamples":[{"configuration":"text-to-video · 1344×768","price":"$0.161","amount":0.1605,"latencyMs":7532,"exampleId":"equine-farrier-service-social-ad"},{"configuration":"first-frame · 768p","price":"$0.161","amount":0.1605,"latencyMs":16368,"exampleId":"limited-edition-welding-helmet-reel"},{"configuration":"text-to-video · 1344×768","price":"$0.161","amount":0.1605,"latencyMs":12790,"exampleId":"volcanic-observatory-documentary-teaser"},{"configuration":"reference-to-video · 768p","price":"$0.709","amount":0.7094,"latencyMs":50378,"exampleId":"municipal-archive-thriller-teaser"}],"guides":[{"slug":"prompting","title":"Writing long, specified prompts","description":"How to write prompts for HeyGen Video 1.0, from naming the camera and ruling out the model's defaults to staging contact, short lettering and prompt enhancement.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/prompting"},{"slug":"dialogue-and-sound","title":"Dialogue, ambience and sound effects","description":"How to direct HeyGen Video 1.0's sound: dialogue sized to the clip length, two speakers, other languages, ambience by distance, effects on cue and music beds.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/dialogue-and-sound"},{"slug":"visual-styles","title":"Animated and illustrated styles","description":"How to take HeyGen Video 1.0 out of live action by naming a production process: stop-motion, cel animation, marker renderings, watercolor, oil paint and screen print.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/visual-styles"},{"slug":"image-to-video","title":"Animating a still with a first frame","description":"How to start a HeyGen Video 1.0 clip from your own image with inputs.frameImages, why the still decides the output shape, and how to get several motions from one packshot.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/image-to-video"},{"slug":"audio-references","title":"Voices and soundtracks from audio references","description":"How to use inputs.referenceAudios with HeyGen Video 1.0 to give a character a specific voice, play your own voiceover word for word, or score a clip with your own track.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/audio-references"},{"slug":"reference-driven-video","title":"Keeping products, people and places with reference images","description":"How to carry a product, a person or a location into a HeyGen Video 1.0 shot with inputs.referenceImages, address each one by label, and control the output shape.","date":"2026-09-30","url":"https://runware.ai/docs/models/heygen-video-1-0/guides/reference-driven-video"}]},{"model":"ideogram-4-5","air":"ideogram:4.5@0","name":"Ideogram 4.5","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-09-30T00:00:00Z","weight":5,"headline":"Text-to-image generation and multi-image editing built for drift-free iteration","description":"Ideogram 4.5 is Ideogram's image generation and editing model, built for iterative work that keeps untouched regions intact across repeated edits. It generates images from text at 1K and 2K presets, and edits a source image using up to four more images as references, with an optional mask to limit where changes land. It suits product imagery, campaign variations, interior restyling, and pose changes guided by a skeleton reference.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-4-5.jpg","ogImage":"https://assets.runware.ai/og/ideogram-4-5.png","pricingRates":[{"amount":0.008,"unit":"output","label":"Very low"},{"amount":0.03,"unit":"output","label":"Low"},{"amount":0.06,"unit":"output","label":"Medium"},{"amount":0.2,"unit":"output","label":"High"}],"pricingExamples":[{"configuration":"text-to-image · 2048×2048","price":"$0.20","amount":0.2,"latencyMs":25717,"exampleId":"regional-polka-album-cover"},{"configuration":"inpaint","price":"$0.20","amount":0.2,"latencyMs":26193,"exampleId":"marine-archaeologist-linocut-portrait"},{"configuration":"reference-to-image · 1536×1024","price":"$0.20","amount":0.2,"latencyMs":32008,"exampleId":"nocturnal-moth-count-social-post"},{"configuration":"text-to-image · 2048×2048","price":"$0.20","amount":0.2,"latencyMs":23619,"exampleId":"industrial-ambient-album-cover"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Ideogram 4.5: the order of a good prompt, space for copy, Magic Prompt, the 1K and 2K sizes and the quality tiers.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5/guides/prompting"},{"slug":"text-and-design","title":"Text and design","description":"How to use Ideogram 4.5 for designs where the words have to be right: quoting exact copy, type hierarchy, placement, dense layouts, non-Latin scripts and brand palettes.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5/guides/text-and-design"},{"slug":"reference-editing","title":"Editing with reference images","description":"How to edit an image with Ideogram 4.5 and up to four reference images: numbering the inputs, taking one attribute from each, choosing the output size and adding a mask.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5/guides/reference-editing"},{"slug":"pose-control","title":"Pose control","description":"How to repose a model photo with Ideogram 4.5 using a pose skeleton reference and joint angles in the prompt, from one raised arm to a full lookbook from a single shot.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5/guides/pose-control"}]},{"model":"ideogram-4-5-precise-edit","air":"ideogram:4.5@precise-edit","name":"Ideogram 4.5 Precise Edit","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-09-30T00:00:00Z","weight":5,"headline":"Instruction-based image editing that restores every pixel the edit leaves untouched","description":"Ideogram 4.5 Precise Edit is the precision editing variant of Ideogram 4.5. It applies a natural-language instruction to a source image, optionally limited by a mask and guided by up to four reference images, and restores unchanged pixels from the source so framing, color, and surrounding detail stay intact across many rounds of edits. It keeps the dimensions of sources up to 4 megapixels, and an edit limited to one region keeps sources up to 8192x8192 at full resolution. It suits retouching, product and color changes, and multi-step design revisions.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-4-5-precise-edit.jpg","ogImage":"https://assets.runware.ai/og/ideogram-4-5-precise-edit.png","pricingRates":[{"amount":0.008,"unit":"output","label":"Very low"},{"amount":0.03,"unit":"output","label":"Low"},{"amount":0.06,"unit":"output","label":"Medium"},{"amount":0.2,"unit":"output","label":"High"}],"pricingExamples":[{"configuration":"image-to-image","price":"$0.20","amount":0.2,"latencyMs":30232,"exampleId":"historic-glasshouse-visitor-free-hero-image"},{"configuration":"reference-to-image","price":"$0.20","amount":0.2,"latencyMs":39578,"exampleId":"heritage-railcar-interior-cleanup"},{"configuration":"inpaint","price":"$0.20","amount":0.2,"latencyMs":36153,"exampleId":"offshore-survival-suit-night-relight"},{"configuration":"image-to-image","price":"$0.20","amount":0.2,"latencyMs":28158,"exampleId":"municipal-telephone-exchange-photo-restoration"},{"configuration":"local-editing","price":"$0.20","amount":0.2,"latencyMs":43095,"exampleId":"apiary-smoker-colorway-variant"},{"configuration":"auto-region-inpaint","price":"$0.20","amount":0.2,"latencyMs":38401,"exampleId":"underwater-camera-housing-colorway"}],"guides":[{"slug":"masked-edits","title":"Masked edits","description":"How to limit an Ideogram 4.5 Precise Edit to one region with a mask, from picking one object out of several to printing a reference graphic onto a garment.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5-precise-edit/guides/masked-edits"},{"slug":"text-localization","title":"Text and localization","description":"How to change and translate the text in a finished design with Ideogram 4.5 Precise Edit, from one updated line to the same banner in four languages.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5-precise-edit/guides/text-localization"},{"slug":"iterative-editing","title":"Iterative editing","description":"How to run a chain of edits with Ideogram 4.5 Precise Edit, one change per round, while every pixel the instruction leaves alone is restored from the source.","date":"2026-09-30","url":"https://runware.ai/docs/models/ideogram-4-5-precise-edit/guides/iterative-editing"},{"slug":"region-edits","title":"Region edits","description":"How to edit one region of a large image with Ideogram 4.5 Precise Edit and settings.editRegion, so 4K and 8K sources come back at full resolution.","date":"2026-10-29","url":"https://runware.ai/docs/models/ideogram-4-5-precise-edit/guides/region-edits"}]},{"model":"openai-gpt-6-1-sol","air":"openai:gpt@6.1-sol","name":"GPT-6.1 Sol","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-09-29T00:00:00Z","weight":5,"headline":"Upgraded reasoning model for coding, computer use, and professional work","description":"GPT-6.1 Sol is OpenAI's upgraded Sol reasoning model for coding, computer use, and professional work. It improves on GPT-6 Sol in writing and debugging code, understanding documents, running multi-step business workflows, and factual accuracy, and combines a 1,050,000-token context window with up to 128,000 output tokens, image understanding, structured outputs, function calling, and broad tool support. Configurable reasoning from low through max suits software engineering, document-heavy analysis, and long-running agents.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-6-1-sol.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-6-1-sol.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"≤ 272K context"},{"amount":0.000004,"unit":"inputToken","label":"> 272K context"},{"amount":0.00001,"unit":"outputToken","label":"≤ 272K context"},{"amount":0.000015,"unit":"outputToken","label":"> 272K context"},{"amount":1e-7,"unit":"cachedInputToken","label":"≤ 272K context"},{"amount":2e-7,"unit":"cachedInputToken","label":"> 272K context"},{"amount":0.0000025,"unit":"cacheWriteToken","label":"≤ 272K context"},{"amount":0.000005,"unit":"cacheWriteToken","label":"> 272K context"}]},{"model":"anthropic-claude-sonnet-5-5","air":"anthropic:claude@sonnet-5.5","name":"Claude Sonnet 5.5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-09-28T00:00:00Z","weight":5,"headline":"Fast, cost-efficient Claude model for everyday coding, agents, and knowledge work","description":"Claude Sonnet 5.5 is the speed and cost tier of Anthropic’s Claude 5.5 family, a faster and lower-cost alternative to Opus 5.5 for well-scoped everyday work. It reads text and images, calls tools, and uses adaptive thinking with configurable effort, with strong results in bug fixing, agentic and computer-use tasks, and polished documents. Its 1M-token context window and 128K-token maximum output cover long conversations, large documents, and multi-file code work.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-sonnet-5-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-sonnet-5-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken"},{"amount":0.00001,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"},{"amount":0.0000025,"unit":"cacheWriteToken"}]},{"model":"bytedance-seedream-5-0-flash","air":"bytedance:seedream@5.0-flash","name":"Seedream 5.0 Flash","creator":"bytedance","status":"live","releasedAt":"2026-09-28T00:00:00Z","weight":5,"headline":"Fast, cost-efficient image generation and multi-reference editing with layer decomposition","description":"Seedream 5.0 Flash is the fast, cost-efficient model in ByteDance's Seedream 5.0 family for image generation and editing. It supports text-to-image and reference-guided editing with up to 10 input images, precise local edits targeted with point and bounding-box coordinates in the prompt, transparent-background output, and decomposition of an image into separate transparent layers. It outputs from 1K to 2K and suits high-volume workflows such as ad concepts, thumbnails, social visuals, and rapid design iteration.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-5-0-flash.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-5-0-flash.png","pricingRates":[{"amount":0.01926,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.0193","amount":0.01926,"latencyMs":18880,"exampleId":"subsea-cable-repair-blog-hero"},{"configuration":"transparent-background · 1024×1024","price":"$0.0193","amount":0.01926,"latencyMs":20338,"exampleId":"fresnel-lens-museum-wall-art"},{"configuration":"reference-to-image · 1024×1536","price":"$0.0193","amount":0.01926,"latencyMs":19076,"exampleId":"forensic-entomologist-profile-avatar"},{"configuration":"layer-decomposition · 2K","price":"$0.193","amount":0.1926,"latencyMs":86919,"exampleId":"kissaten-listening-room-wall-print"}]},{"model":"openai-gpt-6-luna","air":"openai:gpt@6-luna","name":"GPT-6 Luna","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-09-22T00:00:00Z","weight":5,"headline":"Efficient reasoning model for focused, high-volume text and vision workloads","description":"GPT-6 Luna is OpenAI's efficient reasoning model for focused, high-volume workloads. It supports a 1,050,000-token context window, up to 128,000 output tokens, image understanding, structured outputs, function calling, and broad tool use through the Responses API. Configurable reasoning from none through max makes it a strong fit for classification, extraction, summarization, routing, workflow automation, and lightweight agent tasks.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-6-luna.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-6-luna.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":1e-7,"unit":"inputToken","label":"≤ 272K context"},{"amount":2e-7,"unit":"inputToken","label":"> 272K context"},{"amount":5e-7,"unit":"outputToken","label":"≤ 272K context"},{"amount":7.5e-7,"unit":"outputToken","label":"> 272K context"},{"amount":1e-8,"unit":"cachedInputToken","label":"≤ 272K context"},{"amount":2e-8,"unit":"cachedInputToken","label":"> 272K context"},{"amount":1.25e-7,"unit":"cacheWriteToken","label":"≤ 272K context"},{"amount":2.5e-7,"unit":"cacheWriteToken","label":"> 272K context"}]},{"model":"openai-gpt-6-sol","air":"openai:gpt@6-sol","name":"GPT-6 Sol","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-09-22T00:00:00Z","weight":5,"headline":"Reasoning model for complex coding and agentic workflows at production scale","description":"GPT-6 Sol is OpenAI's reasoning model for complex coding and agentic workflows. It combines a 1,050,000-token context window with up to 128,000 output tokens, image understanding, structured outputs, function calling, and broad tool support through the Responses API. Configurable reasoning from none through max makes it suited to software engineering, multi-step automation, technical analysis, and long-running agents that must plan, execute, and verify their work.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-6-sol.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-6-sol.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"≤ 272K context"},{"amount":0.000004,"unit":"inputToken","label":"> 272K context"},{"amount":0.00001,"unit":"outputToken","label":"≤ 272K context"},{"amount":0.000015,"unit":"outputToken","label":"> 272K context"},{"amount":2e-7,"unit":"cachedInputToken","label":"≤ 272K context"},{"amount":4e-7,"unit":"cachedInputToken","label":"> 272K context"},{"amount":0.0000025,"unit":"cacheWriteToken","label":"≤ 272K context"},{"amount":0.000005,"unit":"cacheWriteToken","label":"> 272K context"}]},{"model":"anthropic-claude-opus-5-5","air":"anthropic:claude@opus-5.5","name":"Claude Opus 5.5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-09-21T00:00:00Z","weight":5,"headline":"Top-capability Claude model for long-horizon agents, complex coding, and deep reasoning","description":"Claude Opus 5.5 is Anthropic’s most capable Opus release, built for agent systems that run for hours, large-codebase engineering, and analysis that spans many documents. It reads text and images, uses tools, and exposes configurable effort and extended thinking, so a team can trade latency against depth on a per-request basis. Its 1M-token context window and 128K-token maximum output suit codebase-wide refactors, research synthesis, and long transcript work.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-opus-5-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-opus-5-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000004,"unit":"inputToken"},{"amount":0.00002,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"},{"amount":0.000005,"unit":"cacheWriteToken"}]},{"model":"xai-grok-4-7","air":"xai:grok@4.7","name":"Grok 4.7","creator":"xai","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-09-21T00:00:00Z","weight":5,"headline":"Frontier reasoning model for coding, agentic tasks, and knowledge work","description":"Grok 4.7 is xAI's frontier model for coding, agentic tasks, and knowledge work, built to work longer on difficult tasks and check its own work more carefully. It reads text and images, calls tools, and offers reasoning effort from low to xhigh, with a 500,000-token context window and no fixed output limit. It suits software engineering, multi-step agents, and document-heavy analysis.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/xai-grok-4-7.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-4-7.png","contextWindow":500000,"maxOutputTokens":500000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"< 200K context"},{"amount":0.000004,"unit":"inputToken","label":"≥ 200K context"},{"amount":0.000006,"unit":"outputToken","label":"< 200K context"},{"amount":0.000012,"unit":"outputToken","label":"≥ 200K context"},{"amount":5e-7,"unit":"cachedInputToken","label":"< 200K context"},{"amount":0.000001,"unit":"cachedInputToken","label":"≥ 200K context"}]},{"model":"alibaba-qwen-image-2-1-pro","air":"alibaba:qwen-image@2.1-pro","name":"Qwen-Image-2.1-Pro","creator":"alibaba","status":"live","releasedAt":"2026-09-20T00:00:00Z","weight":5,"headline":"Unified image generation and multi-reference editing with LLM prompt rewriting","description":"Qwen-Image-2.1-Pro is the Pro tier of Alibaba's Qwen-Image-2.1 family, a unified model for text-to-image generation and prompt-guided image editing. It composes and edits from up to 10 reference images and can rewrite prompts with an LLM before generation, in a direct or agent mode with optional thinking. It outputs from 0.26 to 4.19 megapixels at aspect ratios from 1:8 to 8:1, suited to product imagery, marketing visuals, and multi-image compositions.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-2-1-pro.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-2-1-pro.png","pricingRates":[{"amount":0.075,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 1024×1536","price":"$0.075","amount":0.075,"latencyMs":59450,"exampleId":"sheep-shearing-docuseries-poster"},{"configuration":"text-to-image · 1536×1024","price":"$0.075","amount":0.075,"latencyMs":69651,"exampleId":"abyssal-cable-repair-game-key-art"},{"configuration":"reference-to-image · 1024×1536","price":"$0.075","amount":0.075,"latencyMs":27622,"exampleId":"antique-automaton-auction-cleanup"},{"configuration":"text-to-image · 1536×1024","price":"$0.075","amount":0.075,"latencyMs":63384,"exampleId":"airport-falconer-editorial-portrait"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Qwen-Image-2.1-Pro, from sizing inside its pixel budget and banner ratios to layering a scene, quoting on-image copy, and locking a result with a seed.","date":"2026-09-30","url":"https://runware.ai/docs/models/alibaba-qwen-image-2-1-pro/guides/prompting"},{"slug":"prompt-extension","title":"Prompt extension","description":"How promptExtend works on Qwen-Image-2.1-Pro: what the LLM rewrite adds, how direct and agent differ, where thinking fits, and why an extended prompt breaks seeds.","date":"2026-09-30","url":"https://runware.ai/docs/models/alibaba-qwen-image-2-1-pro/guides/prompt-extension"},{"slug":"editing-images","title":"Editing images","description":"How to edit a single image with Qwen-Image-2.1-Pro: naming what changes and what holds, swapping label copy, circling the region to edit, and chaining edits.","date":"2026-09-30","url":"https://runware.ai/docs/models/alibaba-qwen-image-2-1-pro/guides/editing-images"},{"slug":"multi-reference-composition","title":"Composing from multiple reference images","description":"How to combine up to ten reference images with Qwen-Image-2.1-Pro: assigning each a role, dressing a model from packshots, staging a room, and building a team photo.","date":"2026-09-30","url":"https://runware.ai/docs/models/alibaba-qwen-image-2-1-pro/guides/multi-reference-composition"},{"slug":"transparent-backgrounds","title":"Transparent backgrounds","description":"How to generate cut-outs with a real alpha channel on Qwen-Image-2.1-Pro: the prompt wording that turns it on, the output formats that keep it, and subject extraction from a photo.","date":"2026-10-05","url":"https://runware.ai/docs/models/alibaba-qwen-image-2-1-pro/guides/transparent-backgrounds"}]},{"model":"openai-gpt-6-astra","air":"openai:gpt@6-astra","name":"GPT-6 Astra","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-09-18T00:00:00Z","weight":5,"headline":"Frontier reasoning and tool-use model for demanding end-to-end professional workflows","description":"GPT-6 Astra is OpenAI's frontier model for complex reasoning, coding, computer use, research, and document creation. It combines a 1,050,000-token context window with up to 128,000 output tokens, image understanding, structured outputs, function calling, and broad tool support. Configurable reasoning from low through max makes it well suited to long-running agent workflows that must plan, execute, and verify work across code, browsers, and professional software.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-6-astra.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-6-astra.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.00001,"unit":"inputToken","label":"≤ 272K context"},{"amount":0.00002,"unit":"inputToken","label":"> 272K context"},{"amount":0.00005,"unit":"outputToken","label":"≤ 272K context"},{"amount":0.000075,"unit":"outputToken","label":"> 272K context"},{"amount":0.000001,"unit":"cachedInputToken","label":"≤ 272K context"},{"amount":0.000002,"unit":"cachedInputToken","label":"> 272K context"},{"amount":0.0000125,"unit":"cacheWriteToken","label":"≤ 272K context"},{"amount":0.000025,"unit":"cacheWriteToken","label":"> 272K context"}]},{"model":"laya","air":"runware:laya@1","name":"Laya","hosted":true,"endpoints":["decisions"],"status":"compatible-only","releasedAt":"2026-09-18T00:00:00Z","weight":5,"headline":"Open-weight System 1 model for low-latency typed decisions without free-form generation","description":"Laya is an open-weight, non-autoregressive System 1 decision model. Its 421M-parameter English checkpoint pairs a ModernBERT-large encoder with a decision head that evaluates a shared text or JSON state against typed choice, score, and boolean questions in a single forward pass. It returns probabilities and confidence instead of prose, which suits ticket routing, moderation, guardrails, relevance scoring, and other low-latency automation.","capabilities":["io:text-to-text","op:classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/laya.jpg","ogImage":"https://assets.runware.ai/og/laya.png","contextWindow":512,"pricingBasis":"token","pricingRates":[{"amount":2e-8,"unit":"inputToken"}]},{"model":"prunaai-p-video-2-pro","air":"prunaai:p-video@2-pro","name":"P-Video-2-Pro","creator":"prunaai","status":"live","releasedAt":"2026-09-17T00:00:00Z","weight":5,"headline":"Cinematic video generation with native audio and first and last frame control","description":"P-Video-2-Pro is Pruna AI's quality-tier video generation model built on MiniMax H3, creating clips from text, from a first frame with an optional last frame, or from up to 9 reference images, 3 reference videos, and 3 reference audio tracks. It handles multi-beat camera moves, heavy physics like water, fire, and fabric, coherent full-body motion, and two-shot dialogue with lip sync, generating audio with every clip. It offers cost, speed, and quality recipes, three levels of prompt expansion, durations from 5 to 15 seconds, and 480p or 768p output at 24 FPS.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-2-pro.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-2-pro.png","pricingRates":[{"amount":0.01,"unit":"durationSecond","label":"From text or frames · 480p · Cost"},{"amount":0.02,"unit":"durationSecond","label":"From text or frames · 480p · Speed"},{"amount":0.04,"unit":"durationSecond","label":"From text or frames · 480p · Quality"},{"amount":0.025,"unit":"durationSecond","label":"From text or frames · 768p · Cost"},{"amount":0.035,"unit":"durationSecond","label":"From text or frames · 768p · Speed"},{"amount":0.075,"unit":"durationSecond","label":"From text or frames · 768p · Quality"},{"amount":0.015,"unit":"durationSecond","label":"From references · 480p · Cost"},{"amount":0.02,"unit":"durationSecond","label":"From references · 480p · Speed"},{"amount":0.025,"unit":"durationSecond","label":"From references · 768p · Cost"},{"amount":0.035,"unit":"durationSecond","label":"From references · 768p · Speed"},{"amount":0.03,"unit":"durationSecond","label":"Reference video · 480p · Cost"},{"amount":0.04,"unit":"durationSecond","label":"Reference video · 480p · Speed"},{"amount":0.06,"unit":"durationSecond","label":"Reference video · 768p · Cost"},{"amount":0.06,"unit":"durationSecond","label":"Reference video · 768p · Speed"},{"amount":0.005,"unit":"inputImage","label":"Images with references · Cost"},{"amount":0.01,"unit":"inputImage","label":"Images with references · Speed"}],"pricingExamples":[{"configuration":"first-frame · 768p · 5s","price":"$0.0904","amount":0.0904,"latencyMs":17686,"exampleId":"tidal-accordion-music-video"},{"configuration":"text-to-video · 768×1344 · 5s","price":"$0.129","amount":0.1292,"latencyMs":24822,"exampleId":"museum-conservator-social-tip"},{"configuration":"reference-to-video · 768p · 10s","price":"$0.373","amount":0.3729,"latencyMs":68904,"exampleId":"midnight-laundromat-spoken-word-video"},{"configuration":"reference-to-video · 768p · 10s","price":"$0.374","amount":0.3742,"latencyMs":76833,"exampleId":"bat-acoustics-logo-identity-sting"},{"configuration":"text-to-video · 1344×768 · 10s","price":"$0.38","amount":0.3797,"latencyMs":82645,"exampleId":"self-rising-flood-barrier-demo"},{"configuration":"text-to-video · 1344×768 · 10s","price":"$0.38","amount":0.3797,"latencyMs":68689,"exampleId":"bronze-foundry-art-pop-music-video"},{"configuration":"first-last-frame · 768p · 15s","price":"$0.539","amount":0.5391,"latencyMs":63223,"exampleId":"cryogenic-seed-vault-b-roll"}],"guides":[{"slug":"prompting","title":"Writing prompts for full shots and sequences","description":"How to write prompts for Pruna P-Video-2-Pro, from a quick exploring prompt to a timed multi-shot brief, and what the three prompt expander levels change.","date":"2026-09-17","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro/guides/prompting"},{"slug":"speed-quality-and-size","title":"Speed, quality, size and length","description":"How to pick cost, speed or quality mode in Pruna P-Video-2-Pro, choose between 480p and 768p, and set a clip length between 5 and 15 seconds.","date":"2026-09-17","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro/guides/speed-quality-and-size"},{"slug":"camera-and-motion","title":"Camera moves, physics and full-body motion","description":"How to direct camera moves in several beats with Pruna P-Video-2-Pro, and how to prompt liquids, fire, fabric and full-body action so they hold together.","date":"2026-09-17","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro/guides/camera-and-motion"},{"slug":"dialogue-and-audio","title":"Dialogue, lip sync and native audio","description":"How to direct the sound Pruna P-Video-2-Pro generates with every clip: the mix, effects timed to the action, quoted lines, two-person dialogue and sung vocals.","date":"2026-09-17","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro/guides/dialogue-and-audio"},{"slug":"first-and-last-frame","title":"Starting and ending on frames you choose","description":"How to animate a still with Pruna P-Video-2-Pro, pin both the first and last frame with inputs.frameImages, and turn a single image into a seamless loop.","date":"2026-09-17","url":"https://runware.ai/docs/models/prunaai-p-video-2-pro/guides/first-and-last-frame"}]},{"model":"typesafe-jev","air":"typesafe:jev@1","name":"Jev 1.13","creator":"typesafe-ai","endpoints":["decisions"],"status":"compatible-only","releasedAt":"2026-09-15T00:00:00Z","weight":5,"headline":"System One decision model for typed choices, scores, and confidence-aware automation","description":"Jev 1.13 is TypeSafe AI's System One decision model for fast, machine-consumable judgments. It evaluates structured or unstructured application state against typed questions and returns bounded choices, ordinal scores, or yes-or-no probabilities with confidence in a parallel pass. Unlike generative language models, Jev does not produce free-form prose; it is designed for routing, classification, ranking, verification, guardrails, and real-time workflows where software must act on predictable outputs.","capabilities":["io:text-to-text","op:classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/typesafe-jev.jpg","ogImage":"https://assets.runware.ai/og/typesafe-jev.png","pricingBasis":"token","pricingRates":[{"amount":4.2e-8,"unit":"inputToken"}]},{"model":"deepseek-v4-1-flash","air":"deepseek:v4.1@flash","name":"DeepSeek-V4.1-Flash","creator":"deepseek","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-09-10T00:00:00Z","weight":5,"headline":"Multimodal 1M-context MoE for efficient coding, reasoning, and long-horizon agent workflows","description":"DeepSeek-V4.1-Flash is a multimodal mixture-of-experts model for coding, reasoning, visual understanding, and tool-using agents. It processes text and images, generates text, and combines a 1M-token context window with continuously adjustable reasoning effort. Its Causal Encoder-Decoder architecture, sparse attention, and compressed KV cache reduce the memory and computation required for long inputs, making it well suited to large-codebase work, document analysis, research, and sustained agentic tasks.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/deepseek-v4-1-flash.jpg","ogImage":"https://assets.runware.ai/og/deepseek-v4-1-flash.png","contextWindow":1048576,"maxOutputTokens":384000,"pricingBasis":"token","pricingRates":[{"amount":1.5e-7,"unit":"inputToken"},{"amount":6e-7,"unit":"outputToken"},{"amount":1e-8,"unit":"cachedInputToken"}]},{"model":"bfl-flux-video-edit-fast","air":"bfl:flux@video-edit-fast","name":"FLUX Video Edit [fast]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-09-09T00:00:00Z","weight":5,"headline":"Fast prompt-driven video editing that preserves the source clip's motion, timing, framing, and audio","description":"FLUX Video Edit [fast] is Black Forest Labs' instruction-based video editing model for changing an existing clip with a natural-language prompt. It can add, remove, or replace objects and characters; alter settings, styles, actions, and on-screen text; and revise or translate dialogue while preserving unedited motion, camera work, timing, framing, and audio. It supports source clips up to 15 seconds and produces 24 FPS output at up to 720p while retaining the original duration and aspect ratio.","capabilities":["io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-video-edit-fast.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-video-edit-fast.png","pricingRates":[{"amount":0.03,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"video-edit","price":"$0.181","amount":0.1813,"latencyMs":44866,"exampleId":"widescreen-ceramic-workshop-outpaint"},{"configuration":"video-edit","price":"$0.181","amount":0.1813,"latencyMs":54317,"exampleId":"municipal-water-documentary-title-sequence"},{"configuration":"video-edit","price":"$0.181","amount":0.1813,"latencyMs":47663,"exampleId":"falconry-exhibition-miniature-reel"},{"configuration":"video-edit","price":"$0.181","amount":0.1813,"latencyMs":178240,"exampleId":"lighthouse-archive-film-restoration"}],"guides":[{"slug":"editing-video","title":"Editing video","description":"How to edit a finished clip with FLUX Video Edit [fast]: one source video, one written instruction, and everything you did not mention arriving untouched.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/editing-video"},{"slug":"source-video","title":"Preparing source video","description":"How to prepare source clips for FLUX Video Edit [fast]: the 15-second, 50 MiB and 720p envelope, what the output inherits, and how to work with footage that does not fit.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/source-video"},{"slug":"changing-the-setting","title":"Changing the setting","description":"How to relight a clip, move it to another season, or swap the background behind a subject with FLUX Video Edit [fast], without regenerating the shot.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/changing-the-setting"},{"slug":"dialogue-and-localization","title":"Dialogue, on-screen text and localization","description":"How to change what someone says with FLUX Video Edit [fast], translate a line into another language, and swap the on-screen text in the same frame.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/dialogue-and-localization"},{"slug":"removing-objects","title":"Removing objects from video","description":"How to take an object, a person or a sticker out of a clip with FLUX Video Edit [fast], naming what continues behind it so the fill holds for every frame.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/removing-objects"},{"slug":"replacing-and-adding","title":"Replacing and adding elements","description":"How to swap a product, change wardrobe, or put something new into a clip with FLUX Video Edit [fast], and why an added object needs its light described.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/replacing-and-adding"},{"slug":"restyling","title":"Restyling a clip","description":"How to convert footage into another medium with FLUX Video Edit [fast], from vector animation to film stock, while the performance and camera move stay exactly as shot.","date":"2026-09-10","url":"https://runware.ai/docs/models/bfl-flux-video-edit-fast/guides/restyling"}]},{"model":"openai-gpt-image-2-5-flare","air":"openai:gpt-image@2.5-flare","name":"GPT-Image-2.5 Flare","creator":"openai","status":"live","releasedAt":"2026-09-08T00:00:00Z","weight":5,"headline":"Fast, high-quality image generation and editing for everyday creative production","description":"GPT-Image-2.5 Flare is OpenAI's fastest model for high-quality image generation and editing, delivering higher-quality images than GPT Image 2 at 50% lower latency. It improves natural lighting, texture detail, reference-subject preservation, localized editing, and consistency across iterative edits, while handling complex layouts and transparent backgrounds. It is designed for creator content, product experiences, visual search, rapid prototyping, and high-volume generation.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-2-5-flare.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-2-5-flare.png","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.00003,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.0228","amount":0.022775,"latencyMs":20009,"exampleId":"portable-microscope-colorway-product-image"},{"configuration":"reference-to-image","price":"$0.0241","amount":0.024089,"latencyMs":19152,"exampleId":"adaptive-rainwear-lookbook-editorial"},{"configuration":"text-to-image","price":"$0.0431","amount":0.043135,"latencyMs":21147,"exampleId":"circular-laundry-campaign-key-visual"},{"configuration":"transparent-background","price":"$0.0471","amount":0.04712,"latencyMs":22767,"exampleId":"forensic-entomologist-exhibit-persona"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt GPT-Image-2.5 Flare: writing a brief it can plan a layout from, pinning counts and positions, and choosing a size from its custom dimension envelope.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/prompting"},{"slug":"quality-levels","title":"Quality levels","description":"How to set rendering effort on GPT-Image-2.5 Flare with settings.quality: what each level changes, what it costs, and which subjects are worth the top of the ladder.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/quality-levels"},{"slug":"editing-images","title":"Editing images","description":"How to edit an image with GPT-Image-2.5 Flare: one source, one written instruction, and the composition, lighting and detail the model holds while it changes what you named.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/editing-images"},{"slug":"subject-consistency","title":"Subject consistency","description":"How to keep one product or character recognisable across new scenes with GPT-Image-2.5 Flare: how many reference images to send, and how to say what each one is for.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/subject-consistency"},{"slug":"masked-editing","title":"Masked editing","description":"How to confine an edit to a painted region with GPT-Image-2.5 Flare: what white and black mean, and why the mask is guidance rather than a hard boundary.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/masked-editing"},{"slug":"iterative-editing","title":"Iterative editing","description":"How to run a chain of edits with GPT-Image-2.5 Flare: feeding each result back as the next source, what accumulates across passes, and when to restate instead of chain.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/iterative-editing"},{"slug":"transparent-backgrounds","title":"Transparent backgrounds","description":"How to generate cut-out assets with GPT-Image-2.5 Flare: settings.background, the output format it forces, and prompting for edges that survive compositing.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-flare/guides/transparent-backgrounds"}]},{"model":"openai-gpt-image-2-5-sunburst","air":"openai:gpt-image@2.5-sunburst","name":"GPT-Image-2.5 Sunburst","creator":"openai","status":"live","releasedAt":"2026-09-08T00:00:00Z","weight":5,"headline":"Premium image generation and editing with precise control across iterative edits","description":"GPT-Image-2.5 Sunburst is OpenAI's most capable model for image generation and editing, designed for premium workflows that need tighter control across detailed and iterative edits. It combines strong reference-subject preservation, localized changes, natural lighting and textures, complex layout handling, and consistent visual direction for production-ready campaign creative, polished product imagery, and other demanding visual design work.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-2-5-sunburst.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-2-5-sunburst.png","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.00003,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.0152","amount":0.015197,"latencyMs":20447,"exampleId":"dance-festival-risograph-portrait"},{"configuration":"transparent-background","price":"$0.0541","amount":0.05414,"latencyMs":40408,"exampleId":"maritime-mystery-streaming-series-poster"},{"configuration":"reference-to-image","price":"$0.0682","amount":0.06822,"latencyMs":33604,"exampleId":"museum-textile-conservator-editorial-portrait"},{"configuration":"text-to-image","price":"$0.167","amount":0.16694,"latencyMs":67608,"exampleId":"industrial-thermal-camera-colorways"}],"guides":[{"slug":"precision-editing","title":"Precision editing","description":"How to use GPT-Image-2.5 Sunburst for detailed and repeated edits: what the longer generation time buys, and when to reach for it instead of Flare.","date":"2026-09-08","url":"https://runware.ai/docs/models/openai-gpt-image-2-5-sunburst/guides/precision-editing"}]},{"model":"minimax-h3-fast","air":"minimax:h3@fast","name":"MiniMax H3 Fast","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-09-08T00:00:00Z","weight":5,"headline":"Fast multimodal video generation with native audio and image, video, and audio reference control","description":"MiniMax H3 Fast is a speed-optimized H3 video model for generating 5 to 15 second clips at 480p or 768p. It supports text-to-video, first-frame image-to-video, and multimodal reference workflows that combine text with video and audio inputs to preserve visual context, motion, or voice characteristics. It is designed for fast previews, batch generation, and interactive creative workflows that need broader reference control than H3 Max Turbo.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-fast.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-fast.png","pricingRates":[{"amount":0.046,"unit":"durationSecond","label":"480p"},{"amount":0.08,"unit":"durationSecond","label":"768p"},{"amount":0.046,"unit":"durationSecond","label":"480p · uploaded video"},{"amount":0.08,"unit":"durationSecond","label":"768p · uploaded video"}],"pricingExamples":[{"configuration":"text-to-video · 480p · 6s","price":"$0.276","amount":0.276,"latencyMs":12853,"exampleId":"couture-apiary-suit-showcase-reel"},{"configuration":"first-frame · 8s","price":"$0.368","amount":0.368,"latencyMs":34427,"exampleId":"pigeon-anthem-music-video-segment"},{"configuration":"first-last-frame · 10s","price":"$0.46","amount":0.46,"latencyMs":25106,"exampleId":"water-rights-documentary-title-sequence"},{"configuration":"reference-to-video · 480p · 12s","price":"$0.884","amount":0.8838,"latencyMs":58501,"exampleId":"freeport-crime-series-title-teaser"}]},{"model":"microsoft-mai-image-2-6","air":"microsoft:mai-image@2.6","name":"MAI-Image-2.6","creator":"microsoft","status":"coming-soon","releasedAt":"2026-09-04T00:00:00Z","weight":5,"headline":"High-quality image generation and precise multi-reference editing for design-ready visual production","description":"MAI-Image-2.6 is Microsoft AI's highest-quality image generation and editing model, built for precise creative control and polished production output. It supports text-to-image generation, controllable edits with up to five reference images, web grounding for current real-world context, and dynamic aspect-ratio selection with output up to 1.5K. Its strengths include text rendering, portraits, photorealism, commercial design, and maintaining people, products, styles, and scenes across complex edits.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/microsoft-mai-image-2-6.jpg","ogImage":"https://assets.runware.ai/og/microsoft-mai-image-2-6.png"},{"model":"google-gemini-3-8-flash","air":"google:gemini@3.8-flash","name":"Gemini 3.8 Flash","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-09-02T00:00:00Z","weight":5,"headline":"Google's most capable Flash model for long-horizon coding, autonomous agents, and complex multimodal work","description":"Gemini 3.8 Flash is Google's most capable model in the Flash family, built for long-horizon software engineering, autonomous agents, and multi-step reasoning across complex professional workflows. It accepts text, images, video, audio, and PDFs, with a 1M-token context window and configurable thinking effort. Its combination of coding strength, tool use, multimodal understanding, and Flash-class efficiency makes it well suited to production assistants and agent systems.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-8-flash.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-8-flash.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":7.5e-7,"unit":"inputToken"},{"amount":0.00000375,"unit":"outputToken"},{"amount":7.5e-8,"unit":"cachedInputToken"}]},{"model":"minimax-h3-max-turbo","air":"minimax:h3@max-turbo","name":"MiniMax H3 Max Turbo","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-09-02T00:00:00Z","weight":5,"headline":"Distilled H3 Max video generation for faster iteration with native audio and first-to-last frame control","description":"MiniMax H3 Max Turbo is a distilled, speed-focused variant of H3 Max designed to preserve its prompt adherence and visual character at higher throughput. It generates 5 to 15 second videos at 480p or 768p from text or an opening image, with optional end-frame guidance for controlled transitions and synchronized audio generated with the video. It is well suited to rapid creative iteration, interactive experiences, and high-volume production workflows.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-max-turbo.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-max-turbo.png","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"480p"},{"amount":0.04,"unit":"durationSecond","label":"768p"}],"pricingExamples":[{"configuration":"first-frame · 768p · 6s","price":"$0.12","amount":0.12,"latencyMs":13301,"exampleId":"textile-conservator-exhibition-briefing"},{"configuration":"text-to-video · 768×1344 · 8s","price":"$0.16","amount":0.16,"latencyMs":34612,"exampleId":"vertical-specialty-mushroom-harvest-b-roll"},{"configuration":"text-to-video · 768×1344 · 8s","price":"$0.16","amount":0.16,"latencyMs":43776,"exampleId":"semiconductor-wafer-transfer-b-roll"},{"configuration":"first-last-frame · 768p · 10s","price":"$0.20","amount":0.2,"latencyMs":25912,"exampleId":"seed-vault-art-pop-music-video"}]},{"model":"anthropic-claude-fable-5-1","air":"anthropic:claude@fable-5.1","name":"Claude Fable 5.1","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-09-01T00:00:00Z","weight":5,"headline":"Frontier multimodal model for long-running coding, research, and professional work","description":"Claude Fable 5.1 is the next release in Anthropic’s Fable line, aimed at long-horizon coding, agentic execution, and multimodal research. It reads text and images, calls tools, and supports extended thinking, with a 1M-token context window and a 128K-token maximum output for work that has to hold an entire project in view. It fits document-heavy analysis, sustained agent runs, and engineering tasks that span many files.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-fable-5-1.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-fable-5-1.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.00001,"unit":"inputToken"},{"amount":0.00005,"unit":"outputToken"},{"amount":2.5e-7,"unit":"cachedInputToken"},{"amount":0.0000125,"unit":"cacheWriteToken"}]},{"model":"google-gemini-omni-flash-1-1","air":"google:gemini@omni-flash-1.1","name":"Gemini Omni Flash 1.1","creator":"google","zdr":true,"status":"deprecated","replacedBy":"google-gemini-omni-flash-1-2","deactivatesAt":"2026-12-03T00:00:00Z","releasedAt":"2026-08-27T00:00:00Z","weight":5,"headline":"4K multimodal video generation and editing with native audio, clip extension, start-to-end interpolation, and reference video control","description":"Gemini Omni Flash 1.1 is Google's updated multimodal video generation and editing model in the Gemini Omni family. It generates native synchronized audio from the prompt, and expands the original Omni Flash workflow with additional 1080p and 4K output modes, a 360p draft mode, scene extension in 3 to 10 second increments up to 30 seconds total, start-to-end frame interpolation for fluid transitions, and reference-to-video generation guided by both images and short video clips. It is built for teams that need stronger continuity, higher-resolution delivery, and more controllable multi-input video creation than the first Omni Flash release.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/google-gemini-omni-flash-1-1.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-omni-flash-1-1.png","pricingBasis":"token","pricingRates":[{"amount":0.0000015,"unit":"inputToken"},{"amount":0.000009,"unit":"outputToken","label":"Text output"},{"amount":0.0000175,"unit":"outputToken","label":"Video output"}],"pricingExamples":[{"configuration":"first-frame · 5s","price":"$0.503","amount":0.503358,"latencyMs":39007,"exampleId":"manuscript-scanner-product-demo"},{"configuration":"reference-to-video · 7s","price":"$0.729","amount":0.7289635,"latencyMs":57394,"exampleId":"arctic-seed-vault-winter-relight"},{"configuration":"text-to-video · 6s","price":"$0.903","amount":0.903048,"latencyMs":69752,"exampleId":"wildfire-meteorologist-video-headshot"},{"configuration":"video-extend · 5s","price":"$1.15","amount":1.1517085,"latencyMs":84326,"exampleId":"sewer-robot-environment-swap"},{"configuration":"first-last-frame · 8s","price":"$1.20","amount":1.203735,"latencyMs":83461,"exampleId":"theatrical-wig-artisan-headshot-retouch"},{"configuration":"video-edit · 6s","price":"$1.85","amount":1.8524685,"latencyMs":119483,"exampleId":"luxury-apiary-suit-showcase-reel"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Gemini Omni Flash 1.1 for video: the five-element structure, camera vocabulary, holding a single shot, directing audio, and the sampling controls.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/prompting"},{"slug":"resolution-and-duration","title":"Resolution and duration","description":"How to size and time a Gemini Omni Flash 1.1 video: the four resolution presets, the supported width and height pairs, and how tier and duration drive what a clip costs.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/resolution-and-duration"},{"slug":"first-and-last-frame","title":"First and last frame control","description":"How to pin the opening and closing frames of a Gemini Omni Flash 1.1 clip with inputs.frameImages, and prompt the transition the model builds between them.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/first-and-last-frame"},{"slug":"reference-driven-video","title":"Reference images and videos","description":"How to carry a character, product, or style into a Gemini Omni Flash 1.1 shot with inputs.referenceImages and inputs.referenceVideos, and when to combine the two.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/reference-driven-video"},{"slug":"editing-video","title":"Editing video","description":"How to edit an existing clip with Gemini Omni Flash 1.1: relighting, weather, and restyling through inputs.video, why short prompts win, and how to chain edits.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/editing-video"},{"slug":"extending-video","title":"Extending a video","description":"How to extend a clip past the 10-second ceiling with Gemini Omni Flash 1.1: sending duration with inputs.video, prompting continuation, and chaining extensions.","date":"2026-08-31","url":"https://runware.ai/docs/models/google-gemini-omni-flash-1-1/guides/extending-video"}]},{"model":"zai-glm-5-3-flash","air":"zai:glm@5.3-flash","name":"GLM-5.3-Flash","creator":"zai","hosted":true,"precision":"fp8","zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-08-26T00:00:00Z","weight":5,"headline":"Efficient multimodal GLM for coding, agents, and long-context visual understanding","description":"GLM-5.3-Flash is Z.ai's lower-cost multimodal model for coding, long-horizon agent workflows, and visual reasoning. It is the first natively multimodal model in the GLM-5 series, combining text, image, and video understanding with stronger coding and agent performance than earlier GLM flash-tier behavior would suggest. Z.ai positions it as a default model for frequent professional work that needs fast feedback, efficient long-context serving, and reliable reasoning across software, documents, charts, interfaces, screenshots, and other visually grounded tasks.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/zai-glm-5-3-flash.jpg","ogImage":"https://assets.runware.ai/og/zai-glm-5-3-flash.png","contextWindow":1048572,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":1.5e-7,"unit":"inputToken"},{"amount":5e-7,"unit":"outputToken"},{"amount":3e-8,"unit":"cachedInputToken"}]},{"model":"minimax-h3-max","air":"minimax:h3@max","name":"MiniMax H3 Max","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-08-26T00:00:00Z","weight":5,"headline":"High-throughput video generation with stronger prompt adherence, native synced audio, and first-to-last frame control","description":"MiniMax H3 Max is a performance-tuned variant of MiniMax H3 built for faster video generation while preserving strong prompt following and polished audiovisual results. It supports text-to-video and image-to-video workflows, including optional first-and-last-frame guidance for more controlled motion between key images, and is well suited to rapid ideation, high-volume content production, and interactive creative workflows that need lower latency without giving up visual quality.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/minimax-h3-max.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3-max.png","pricingRates":[{"amount":0.05,"unit":"durationSecond","label":"480p"},{"amount":0.08,"unit":"durationSecond","label":"768p"}],"pricingExamples":[{"configuration":"first-frame · 768p · 5s","price":"$0.40","amount":0.4,"latencyMs":23476,"exampleId":"vineyard-falconry-social-video-ad"},{"configuration":"first-last-frame · 480p · 15s","price":"$0.75","amount":0.75,"latencyMs":25360,"exampleId":"accordion-restoration-reel-ad"},{"configuration":"text-to-video · 1344×768 · 10s","price":"$0.80","amount":0.8,"latencyMs":23978,"exampleId":"mycelium-biomaterial-lab-b-roll"},{"configuration":"text-to-video · 1344×768 · 10s","price":"$0.80","amount":0.8,"latencyMs":23452,"exampleId":"museum-seismic-plinth-explainer"}],"guides":[{"slug":"generating-video","title":"Generating video","description":"How to generate video with MiniMax H3 Max: the fixed size pairs at 480p and 768p, animating a first and last frame, native synced audio, and rendering faster than playback.","date":"2026-08-31","url":"https://runware.ai/docs/models/minimax-h3-max/guides/generating-video"}]},{"model":"alibaba-wan3-0","air":"alibaba:wan@3.0","name":"Wan3.0","creator":"alibaba","status":"live","releasedAt":"2026-08-24T00:00:00Z","weight":5,"headline":"All-in-one multimodal video generation with native 30-second clips, large reference capacity, and precise video editing","description":"Wan3.0 is Alibaba's higher-end all-in-one multimodal video model for longer, reference-heavy, production-oriented creation. It combines text-to-video, first- and last-frame image-to-video, reference-driven video generation, localized video editing, and temporal extension in one model, with support for image, video, audio, document, and webpage inputs. It is especially well suited to branded storytelling, product demos, explainers, music-led sequences, and other commercial workflows that need stronger character or product consistency, more impactful audiovisual motion, and tighter control over complex multi-input prompts.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan3-0.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan3-0.png","pricingRates":[{"amount":0.05,"unit":"durationSecond","label":"480p"},{"amount":0.1,"unit":"durationSecond","label":"720p"},{"amount":0.2,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s","price":"$0.504","amount":0.5038,"latencyMs":127753,"exampleId":"mycelium-packaging-factory-b-roll"},{"configuration":"text-to-video · 720p · 6s · audio","price":"$0.604","amount":0.6037,"latencyMs":185663,"exampleId":"restaurant-knife-sharpening-service-reel"},{"configuration":"url-grounded · 960×960 · 7s","price":"$0.704","amount":0.7035,"latencyMs":523292,"exampleId":"climbing-helmet-colorway-reveal"},{"configuration":"first-last-frame · 1080p · 8s","price":"$1.62","amount":1.616,"latencyMs":192264,"exampleId":"airport-falconry-team-introduction"},{"configuration":"reference-to-video · 1080p · 10s","price":"$2.01","amount":2.0062,"latencyMs":423554,"exampleId":"hospital-tube-network-documentary-teaser"}],"guides":[{"slug":"prompting","title":"Prompting Wan 3.0","description":"How to write prompts for Wan 3.0 that get you the shot: how the model reads layered directives, camera language, style and register range, directing native audio, and the dimensions and duration rules.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-wan3-0/guides/prompting"},{"slug":"image-inputs","title":"Image inputs with Wan 3.0","description":"How to use images with Wan 3.0's video generation: pin an image as the first frame, morph between two frames, and carry subject or product identity across new scenes with reference images.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-wan3-0/guides/image-inputs"},{"slug":"video-and-audio-references","title":"Video and audio references with Wan 3.0","description":"How to use video and audio references with Wan 3.0: carrying a subject across new scenes with referenceVideos, driving video from an audio source with referenceAudios, and combining multiple reference types in one call.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-wan3-0/guides/video-and-audio-references"},{"slug":"document-and-web-inputs","title":"Document and web inputs with Wan 3.0","description":"How to turn structured content into video with Wan 3.0: recipes as cooking videos and public webpages as short landmark or historical montages.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-wan3-0/guides/document-and-web-inputs"}]},{"model":"alibaba-wan3-0-prime","air":"alibaba:wan@3.0-prime","name":"Wan3.0 Prime","creator":"alibaba","status":"live","releasedAt":"2026-08-24T00:00:00Z","weight":5,"headline":"Lower-latency Wan3.0 video generation with the same multimodal workflows and output quality","description":"Wan3.0 Prime is the faster inference variant of Alibaba's Wan3.0 video model. It keeps the same API shape, the same multimodal creation and editing workflows, and the same output quality as Wan3.0, while trading up to a higher price tier for lower latency. It supports text-to-video, first- and last-frame image-to-video, reference-driven video generation, localized video editing, and temporal extension, with support for image, video, audio, document, and webpage inputs in the broader Wan3.0 workflow.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan3-0-prime.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan3-0-prime.png","pricingRates":[{"amount":0.068,"unit":"durationSecond","label":"480p"},{"amount":0.14,"unit":"durationSecond","label":"720p"},{"amount":0.28,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"url-grounded · 960×960 · 4s","price":"$0.566","amount":0.5656,"latencyMs":536689,"exampleId":"thermal-camera-safety-colorway-reel"},{"configuration":"first-frame · 720p · 5s","price":"$0.705","amount":0.7053,"latencyMs":49600,"exampleId":"rare-seed-vault-retrieval-b-roll"},{"configuration":"reference-to-video · 832×1104 · 6s","price":"$0.845","amount":0.8452,"latencyMs":99183,"exampleId":"watchmaker-demagnetizer-product-demo"},{"configuration":"text-to-video · 1080p · 6s · audio","price":"$1.69","amount":1.6904,"latencyMs":253462,"exampleId":"archival-market-widescreen-expansion"},{"configuration":"first-last-frame · 1080p · 7s","price":"$1.98","amount":1.983,"latencyMs":462474,"exampleId":"ballast-compliance-avatar-briefing"}]},{"model":"zai-glm-5-3","air":"zai:glm@5.3","name":"GLM-5.3","creator":"zai","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-08-14T00:00:00Z","weight":5,"headline":"Flagship coding and agentic LLM with stronger long-horizon execution and security-oriented reasoning","description":"GLM-5.3 is Z.ai's flagship language model for coding, tool-using agents, and sustained multi-step problem solving. It builds on the long-horizon engineering focus of the recent GLM line with stronger software engineering performance, better benchmark results across terminal and agent evaluations, and notably stronger capability in security-oriented code analysis and vulnerability discovery. It is a strong fit for demanding coding assistants, automation systems, and engineering workflows that need reliable reasoning over large task context.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/zai-glm-5-3.jpg","ogImage":"https://assets.runware.ai/og/zai-glm-5-3.png","contextWindow":1048576,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":0.0000012,"unit":"inputToken"},{"amount":0.000004,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"}]},{"model":"recraft-v4-styles-pro-vector","air":"recraft:v4@styles-pro-vector","name":"Recraft V4 Styles Pro Vector","creator":"recraft","status":"live","releasedAt":"2026-08-13T00:00:00Z","weight":5,"headline":"Premium style-consistent SVG generation for high-end brand and design systems","description":"Recraft V4 Styles Pro Vector is the premium vector model in the V4 Styles family. It pairs reusable reference-based style creation with SVG output for workflows that need stronger detail retention and more polished scalable assets. It is well suited to brand systems, icon libraries, packaging graphics, editorial illustration systems, and other vector-first pipelines where a consistent visual identity needs to carry across many generated assets without training a dedicated custom model.","capabilities":["io:text-to-image","io:image-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-styles-pro-vector.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-styles-pro-vector.png","pricingRates":[{"amount":0.125,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.125","amount":0.125,"latencyMs":105183,"exampleId":"wastewater-outbreak-detection-explainer"},{"configuration":"vectorize","price":"$0.125","amount":0.125,"latencyMs":35143,"exampleId":"subsea-cable-crew-shirt-graphic"},{"configuration":"vectorize","price":"$0.125","amount":0.125,"latencyMs":41414,"exampleId":"street-sweeper-recruitment-campaign-illustration"},{"configuration":"vectorize","price":"$0.125","amount":0.125,"latencyMs":40551,"exampleId":"food-spectrometer-lifestyle-illustration"}]},{"model":"xai-grok-imagine-image-2-0","air":"xai:grok-imagine@image-2.0","name":"Grok Imagine Image 2.0","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-08-10T00:00:00Z","weight":5,"headline":"Next-generation Grok image generation and editing with quality-aware output control","description":"Grok Imagine Image 2.0 is xAI's next-generation image generation and editing model for both text-to-image and prompt-guided image transformation. It keeps the same core workflow as the current Grok Imagine image family, including aspect-ratio control and image-based editing, while adding a dedicated quality parameter so teams can tune output fidelity within the same API shape. It is a strong fit for creative production pipelines that want one Grok image endpoint for generation, editing, and quality-sensitive iteration without switching to a separate model family.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/xai-grok-imagine-image-2-0.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-imagine-image-2-0.png","pricingOverview":"Price per generated image with an additional $0.01 per input image for edits. ","pricingRates":[{"amount":0.04,"unit":"output","label":"quality low · 1K"},{"amount":0.06,"unit":"output","label":"quality medium · 1K"},{"amount":0.06,"unit":"output","label":"quality low · 2K"},{"amount":0.08,"unit":"output","label":"quality medium · 2K"},{"amount":0.01,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1248×832 · quality medium","price":"$0.06","amount":0.06,"latencyMs":103974,"exampleId":"underground-cycling-tourism-campaign-image"},{"configuration":"text-to-image · 1248×832 · quality medium","price":"$0.06","amount":0.06,"latencyMs":113874,"exampleId":"concert-harp-auction-cutout"},{"configuration":"reference-to-image · 2K · quality low","price":"$0.08","amount":0.08,"latencyMs":22779,"exampleId":"ammonite-exhibition-gala-invitation"},{"configuration":"reference-to-image · 2K · quality low","price":"$0.08","amount":0.08,"latencyMs":22216,"exampleId":"national-seed-curator-persona-portrait"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write text-to-image prompts for Grok Imagine Image 2.0: structuring a shot, directing the camera and light, the 1K and 2K aspect pairs, and its factual detail.","date":"2026-08-10","url":"https://runware.ai/docs/models/xai-grok-imagine-image-2-0/guides/prompting"},{"slug":"text-rendering","title":"Rendering accurate text","description":"How to render accurate, readable text with Grok Imagine Image 2.0: quoting the exact words, directing typography and hierarchy, and holding up on small, dense layouts.","date":"2026-08-10","url":"https://runware.ai/docs/models/xai-grok-imagine-image-2-0/guides/text-rendering"},{"slug":"editing","title":"Editing images","description":"How to edit an image with Grok Imagine Image 2.0: recolour, restyle, remove objects, swap backgrounds, and relight from one reference image and a prompt.","date":"2026-08-10","url":"https://runware.ai/docs/models/xai-grok-imagine-image-2-0/guides/editing"}]},{"model":"bytedance-seedance-2-5","air":"bytedance:seedance@2.5","name":"Seedance 2.5","creator":"bytedance","status":"live","releasedAt":"2026-08-07T00:00:00Z","weight":5,"headline":"Professional multimodal video generation with native 30-second clips, large-scale reference control, and precise localized editing","description":"Seedance 2.5 is ByteDance's higher-end multimodal video generation model for production-oriented creative work. It supports native 30-second video generation, larger reference sets across image, video, and audio inputs, and stronger instruction following for complex shots, multilingual prompts, and reference-heavy creative direction. It is especially well suited to branded storytelling, advertising, explainers, and other commercial workflows that need longer continuous scenes, tighter subject and style consistency, and precise edits that preserve the rest of the shot.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-2-5.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-2-5.png","pricingOverview":"$10.7/M Token for Text/Image to Video at 480p and 720p; \n$12.52/M Token for Text/Image to Video at 1080p;\n$6.848/M Token for Video-to-Video at 480p and 720p; \n$7.49/M Token for Video-to-Video  at 1080p;\n","pricingRates":[{"amount":0.1025,"unit":"durationSecond","label":"Text/Image to Video · 480p"},{"amount":0.2304,"unit":"durationSecond","label":"Text/Image to Video · 720p"},{"amount":0.61354,"unit":"durationSecond","label":"Text/Image to Video · 1080p"},{"amount":0.1313,"unit":"durationSecond","label":"Video-to-Video · 480p"},{"amount":0.2953,"unit":"durationSecond","label":"Video-to-Video · 720p"},{"amount":0.73415,"unit":"durationSecond","label":"Video-to-Video · 1080p"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 5s · audio","price":"$1.17","amount":1.16523,"latencyMs":143889,"exampleId":"midwifery-archive-restoration-reveal"},{"configuration":"video-edit · 720p · autos","price":"$1.78","amount":1.7811648,"latencyMs":175929,"exampleId":"wildlife-tracking-collar-cleanup"},{"configuration":"reference-to-video · 720p · 8s","price":"$1.86","amount":1.85859,"latencyMs":232972,"exampleId":"rooftop-apiary-scene-replacement"},{"configuration":"first-last-frame · 720p · 10s","price":"$2.32","amount":2.32083,"latencyMs":104832,"exampleId":"four-track-recorder-colorway-reveal-film"},{"configuration":"first-frame · 1080p · 6s","price":"$3.67","amount":3.67441968,"latencyMs":176184,"exampleId":"geothermal-spa-equipment-cleanup"},{"configuration":"video-extend · 1080p · 6s","price":"$4.37","amount":4.368168,"latencyMs":201781,"exampleId":"quarry-photo-linocut-transition"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Seedance 2.5 for directed text-to-video: the five-layer shot scaffold, camera vocabulary the model reads directly, second-level timing, and native audio.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/prompting"},{"slug":"image-to-video","title":"Image to video and keyframes","description":"How to turn a still into video with Seedance 2.5: animate a first frame, interpolate a first and last frame, loop a still, and sequence images as keyframes.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/image-to-video"},{"slug":"multi-reference-production","title":"Multimodal reference","description":"How to compose up to 30 images, 10 videos, and 10 audio clips into one Seedance 2.5 call with typed @-tag addressing, from cast ensembles and product families to audio-driven scenes.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/multi-reference-production"},{"slug":"editing","title":"Editing video","description":"How to edit an existing clip in place with Seedance 2.5 using inputs.video: remove and replace elements, restyle and relight the frame, swap backgrounds, and edit by timestamp.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/editing"},{"slug":"long-form-narrative","title":"Long-form and extension","description":"How to structure a 30-second Seedance 2.5 brief as one continuous take or a multi-shot passage, hold continuity across the runtime, and extend past 30 seconds with inputs.video.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/long-form-narrative"},{"slug":"motion-control","title":"Motion and performance transfer","description":"How to drive Seedance 2.5 with a reference video for motion, camera, and lip-sync while a reference image supplies identity: motion transfer, character swaps, and clay to finished shot.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/motion-control"},{"slug":"multilingual","title":"Multilingual video","description":"How to generate video in 10+ languages with Seedance 2.5: native lip-sync from a text prompt, switching languages within one clip, and controlling accent and delivery.","date":"2026-08-07","url":"https://runware.ai/docs/models/bytedance-seedance-2-5/guides/multilingual"}]},{"model":"bfl-flux-3-video","air":"bfl:flux@3-video","name":"FLUX 3 Video","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-08-04T00:00:00Z","weight":5,"headline":"Multimodal video generation with native synchronized audio across styles and modes","description":"FLUX 3 Video is Black Forest Labs' multimodal foundation model for video generation with synchronized audio. It generates clips from 5 to 20 seconds across text-to-video, image-to-video, and video-to-video modes on one architecture, with keyframe control to pin an opening image or interpolate motion across pinned frames, chained continuations for arcs beyond 20 seconds, multi-shot sequences with hard cuts inside one generation, and native multilingual dialogue. A draft mode returns a fast low-resolution preview and a cache that a follow-up call enhances at full quality, tightening iteration loops. Style range spans candid camcorder footage, animation, motion design, and cinematic photoreal, character consistency holds across scenes within one generation, and in-video typography renders cleanly for titles and animated designs.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-3-video.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-3-video.png","pricingRates":[{"amount":0.04,"unit":"durationSecond","label":"T2V & I2V. Draft · 720p"},{"amount":0.06,"unit":"durationSecond","label":"V2V · Draft · 720p"},{"amount":0.17,"unit":"durationSecond","label":"T2V & I2V. 720p"},{"amount":0.29,"unit":"durationSecond","label":"T2V & I2V · 1080p"},{"amount":0.43,"unit":"durationSecond","label":"V2V · 720p"},{"amount":0.54,"unit":"durationSecond","label":"V2V · 1080p"}],"pricingExamples":[{"configuration":"draft-preview · 720p · 5s","price":"$0.202","amount":0.2017,"latencyMs":155516,"exampleId":"urban-bat-nonprofit-logo-ident"},{"configuration":"first-frame · 720p · 7s","price":"$1.20","amount":1.1971,"latencyMs":87042,"exampleId":"field-water-photometer-demo"},{"configuration":"text-to-video · 720p · 8s · audio","price":"$1.37","amount":1.3671,"latencyMs":142139,"exampleId":"autonomous-pipe-crawler-product-demo"},{"configuration":"multiple-keyframes · 1088×800 · 12s","price":"$2.05","amount":2.0471,"latencyMs":249224,"exampleId":"inflatable-marine-beacon-deployment-demo"},{"configuration":"video-extend · 720p · 6s","price":"$2.58","amount":2.58,"latencyMs":179536,"exampleId":"mycelium-packaging-social-launch-reel"},{"configuration":"first-last-frame · 1080p · 10s","price":"$2.91","amount":2.9121,"latencyMs":226914,"exampleId":"observatory-chart-lyric-video"}],"guides":[{"slug":"prompting","title":"Prompting FLUX 3","description":"How to prompt FLUX 3 for text-to-video with synchronized audio: request shape, prompt rewriting, camera language, dimensions, style diversity, and the draft-mode iteration workflow.","date":"2026-08-04","url":"https://runware.ai/docs/models/bfl-flux-3-video/guides/prompting"},{"slug":"audio-and-speech","title":"Audio and speech with FLUX 3","description":"How to compose FLUX 3's synchronized audio: layering ambient and effects, directing music by instrumentation and tempo, directing speech delivery, and rendering multilingual dialogue with clean lip-sync.","date":"2026-08-04","url":"https://runware.ai/docs/models/bfl-flux-3-video/guides/audio-and-speech"},{"slug":"multi-shot-sequences","title":"Multi-shot sequences with FLUX 3","description":"How to build multi-shot video sequences inside one FLUX 3 generation: HARD CUT syntax, shot contrast, time compression, threading an audio bed across the cuts, and pacing shot rhythm.","date":"2026-08-04","url":"https://runware.ai/docs/models/bfl-flux-3-video/guides/multi-shot-sequences"},{"slug":"keyframes","title":"Keyframes with FLUX 3","description":"How to pin images to specific frame positions in FLUX 3 videos: opening on a source image, storyboarding across positions, ending on a packshot, morphing between two states, and using timestamps for beat-precise timing.","date":"2026-08-04","url":"https://runware.ai/docs/models/bfl-flux-3-video/guides/keyframes"},{"slug":"video-continuation","title":"Video continuation with FLUX 3","description":"How to continue a source clip from its final frames with FLUX 3's video input: single-clip continuation, chaining across multiple generations, deliberate source design, and recovering shots that didn't land.","date":"2026-08-04","url":"https://runware.ai/docs/models/bfl-flux-3-video/guides/video-continuation"}]},{"model":"alibaba-qwen-image-3-0","air":"alibaba:qwen-image@3.0","name":"Qwen-Image-3.0","creator":"alibaba","status":"live","releasedAt":"2026-08-04T00:00:00Z","weight":5,"headline":"Unified image generation and editing that balances visual quality, layout fidelity, and response speed","description":"Qwen-Image-3.0 is the standard model in Alibaba's Qwen Image 3.0 family, supporting both text-to-image generation and prompt-guided image editing in a single endpoint. It keeps the family’s strengths in structured visual composition, typography-heavy generation, multilingual text rendering, and multi-image editing, but is positioned as the quality-speed balanced option rather than the highest-fidelity Pro tier. It is a strong fit for everyday creative production where teams want one model for generation and editing, support for one to three reference images, and reliable handling of layout-sensitive prompts without always reaching for the heavier model.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-3-0.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-3-0.png","pricingOverview":"$0.03 per image output, with an additional $0.003 per input image","pricingRates":[{"amount":0.03,"unit":"output"},{"amount":0.003,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.03","amount":0.03,"latencyMs":79086,"exampleId":"adaptive-beach-wheelchair-lifestyle-hero"},{"configuration":"text-to-image · 1536×1024","price":"$0.03","amount":0.03,"latencyMs":86937,"exampleId":"tidal-turbine-linocut-wall-art"},{"configuration":"reference-to-image · 1024×1536","price":"$0.036","amount":0.036,"latencyMs":89394,"exampleId":"horseshoe-crab-volunteer-social-post"},{"configuration":"reference-to-image · 1024×1536","price":"$0.036","amount":0.036,"latencyMs":76942,"exampleId":"curbside-compost-pickup-pinterest-ad"}],"guides":[{"slug":"prompting","title":"Prompting Qwen Image 3.0","description":"How to prompt Qwen Image 3.0, from sizing inside its pixel budget to layering a scene description, steering with negativePrompt, and locking a result with a seed.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-qwen-image-3-0/guides/prompting"},{"slug":"prompt-extension","title":"Prompt extension with Qwen Image 3.0","description":"How promptExtend works on Qwen Image 3.0: what the LLM rewrite adds to a short prompt, how the direct and agent modes differ, and why extension and seeds don't mix.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-qwen-image-3-0/guides/prompt-extension"},{"slug":"text-and-typography","title":"Text and typography with Qwen Image 3.0","description":"How to render exact copy with Qwen Image 3.0: quoting the strings you need, building a type hierarchy, setting Chinese and bilingual layouts, and sizing for print.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-qwen-image-3-0/guides/text-and-typography"},{"slug":"reference-images","title":"Editing and composing with reference images","description":"How to use reference images on Qwen Image 3.0 to recolor a packshot, carry a product into a new scene, or stage two products together in one shot.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-qwen-image-3-0/guides/reference-images"},{"slug":"dense-layouts","title":"Dense layouts and information design","description":"How to generate infographics, comparison grids, newspaper pages, and interface mockups with Qwen Image 3.0 by spending its long prompt budget on every panel.","date":"2026-08-04","url":"https://runware.ai/docs/models/alibaba-qwen-image-3-0/guides/dense-layouts"}]},{"model":"alibaba-qwen-image-3-0-pro","air":"alibaba:qwen-image@3.0-pro","name":"Qwen-Image-3.0-Pro","creator":"alibaba","status":"live","releasedAt":"2026-08-04T00:00:00Z","weight":5,"headline":"High-control image generation and editing for dense layouts, small text rendering, and multilingual visual design","description":"Qwen-Image-3.0-Pro is Alibaba's higher-capability Qwen Image 3.0 model for both text-to-image generation and prompt-guided image editing. It is built for visually dense and instruction-heavy work, with stronger support for complex layouts, image-within-image compositions, multilingual typography, and precise small-text rendering while preserving high photorealistic detail in faces, skin, hair, and materials. It is especially well suited to posters, menus, storyboards, interface mockups, branded graphics, and other production workflows where layout fidelity and text quality matter as much as raw image quality.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-3-0-pro.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-3-0-pro.png","pricingOverview":"Starts at $0.04 per image output. With an additional $0.003 per input image","pricingRates":[{"amount":0.04,"unit":"output","label":"1K output image"},{"amount":0.075,"unit":"output","label":"2K output image"},{"amount":0.003,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.04","amount":0.04,"latencyMs":11777,"exampleId":"municipal-geothermal-campaign-key-visual"},{"configuration":"text-to-image · 1536×1024","price":"$0.04","amount":0.04,"latencyMs":10828,"exampleId":"colorized-cable-ship-archive-photograph"},{"configuration":"reference-to-image · 1024×1536","price":"$0.046","amount":0.046,"latencyMs":153610,"exampleId":"museum-pest-investigation-blog-hero"},{"configuration":"reference-to-image · 1024×1536","price":"$0.046","amount":0.046,"latencyMs":127289,"exampleId":"atlantic-dulse-carton-packaging"}]},{"model":"deepseek-v4-flash","air":"deepseek:v4@flash","name":"DeepSeek-V4-Flash-0731","creator":"deepseek","hosted":true,"precision":"fp8","zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-07-31T00:00:00Z","weight":5,"headline":"Fast frontier LLM with 1M context, tool use, and dual thinking modes","description":"DeepSeek-V4-Flash is DeepSeek's fast, efficient, and cost-focused frontier language model for coding, reasoning, and agent workflows. It supports both thinking and non-thinking modes, a 1M token context window, up to 384K output tokens, tool calls, JSON output, and efficient long-context operation for software, research, and structured professional tasks.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/deepseek-v4-flash.jpg","ogImage":"https://assets.runware.ai/og/deepseek-v4-flash.png","contextWindow":1048576,"maxOutputTokens":384000,"pricingBasis":"token","pricingRates":[{"amount":7.6e-8,"unit":"inputToken"},{"amount":1.53e-7,"unit":"outputToken"},{"amount":1.4e-8,"unit":"cachedInputToken"}]},{"model":"minimax-h3","air":"minimax:h3@0","name":"MiniMax H3","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-07-30T00:00:00Z","weight":5,"headline":"Multimodal video generation with native synced audio, multi-reference consistency, and continuation workflows","description":"MiniMax H3 is a multimodal video generation model that supports text-to-video, first-frame and keyframe-guided generation, multi-reference conditioning, and audio-video continuation in a single workflow. It accepts text together with images, videos, and audio references to keep subjects, voice, motion, and scene identity more consistent across shots, while generating synchronized sound natively rather than as a separate dubbing pass. It is well suited to cinematic multi-shot generation, reference-driven character performance, instruction-based video editing, and continuation workflows that extend an existing clip or audio segment into a seamless new video.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/minimax-h3.jpg","ogImage":"https://assets.runware.ai/og/minimax-h3.png","pricingOverview":"Output video costs $0.08 per second at 768p and $0.13 per second at 2K. Input video costs $0.13/s of uploaded video. The first 5 input images are free then it is $0.04 for any additional ones. Input audio is free.","pricingRates":[{"amount":0.08,"unit":"durationSecond","label":"768p"},{"amount":0.13,"unit":"durationSecond","label":"2K"},{"amount":0.04,"unit":"inputImage","after":5,"label":"each image beyond the first five"},{"amount":0.13,"unit":"durationSecond","label":"uploaded video"}],"pricingExamples":[{"configuration":"text-to-video · 1344×768 · 6s","price":"$0.48","amount":0.48,"latencyMs":146756,"exampleId":"avalanche-patrol-team-portrait"},{"configuration":"first-frame · 768p · 8s","price":"$0.64","amount":0.64,"latencyMs":174046,"exampleId":"indoor-wasabi-farm-b-roll"},{"configuration":"first-last-frame · 1440p · 10s","price":"$1.30","amount":1.3,"latencyMs":446530,"exampleId":"underwater-scooter-colorway-reveal-film"},{"configuration":"reference-to-video · 768p · 12s","price":"$1.74","amount":1.74,"latencyMs":441866,"exampleId":"wastewater-lab-public-health-briefing"}],"guides":[{"slug":"generating-video","title":"Generating video","description":"How to generate video from text with MiniMax H3: the six 2K aspect ratios, 5 to 15 second durations, native synced audio, and prompting for cinematic shots.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/generating-video"},{"slug":"motion-camera-performance","title":"Motion, camera, and performance","description":"How to transfer motion, a camera move, or an acting performance from a reference video onto a new subject with MiniMax H3 Omni Reference.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/motion-camera-performance"},{"slug":"sound-and-voice","title":"Sound and voice","description":"How to direct MiniMax H3's native audio from the prompt: ambience, synced sound effects, music, and spoken dialogue with lip-sync.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/sound-and-voice"},{"slug":"first-and-last-frame","title":"First and last frame","description":"How to animate a still image with MiniMax H3 and bridge a first and last frame into one continuous shot, with the output following the image's own aspect ratio.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/first-and-last-frame"},{"slug":"reference-driven-consistency","title":"Reference-driven consistency","description":"How to lock a character, product, or style across a new MiniMax H3 shot with Omni Reference images, and address each reference by index in the prompt.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/reference-driven-consistency"},{"slug":"editing-video","title":"Editing video","description":"How to edit a finished clip with a prompt in MiniMax H3: replace a subject, relight a scene, add or remove elements, and combine edits, keeping the rest untouched.","date":"2026-07-30","url":"https://runware.ai/docs/models/minimax-h3/guides/editing-video"}]},{"model":"anthropic-claude-opus-5","air":"anthropic:claude@opus-5","name":"Claude Opus 5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-07-24T00:00:00Z","weight":5,"headline":"High-capability Claude model for deep reasoning, long-horizon agents, and complex coding and enterprise work","description":"Claude Opus 5 is Anthropic's high-capability model for complex agentic coding, deep reasoning, long-running tool use, and enterprise knowledge work. It combines a 1M-token context window with vision, configurable effort levels, and reasoning enabled by default, allowing teams to balance latency and cost against capability. It is particularly suited to large codebases, multi-step research and analysis, document-heavy workflows, and agent systems that need sustained, reliable execution.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-opus-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-opus-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken"},{"amount":0.000025,"unit":"outputToken"},{"amount":5e-7,"unit":"cachedInputToken"},{"amount":0.00000625,"unit":"cacheWriteToken"}]},{"model":"moonshotai-kimi-k3","air":"moonshotai:kimi@k3","name":"Kimi K3","creator":"moonshot-ai","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-07-23T00:00:00Z","weight":5,"headline":"Flagship multimodal reasoning LLM for long-horizon coding, knowledge work, and tool-rich agent workflows","description":"Kimi K3 is Moonshot AI's flagship model for frontier reasoning, software engineering, and knowledge-intensive work. It combines a 1M-token context window with native visual understanding, configurable reasoning effort, structured outputs, tool calling, and strong long-horizon execution across large codebases and multi-step research workflows. It is well suited to demanding coding agents, deep research systems, document-heavy assistants, and other production applications that need sustained reasoning over very large contexts.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/moonshotai-kimi-k3.jpg","ogImage":"https://assets.runware.ai/og/moonshotai-kimi-k3.png","contextWindow":1048576,"maxOutputTokens":1048576,"pricingBasis":"token","pricingRates":[{"amount":0.000003,"unit":"inputToken"},{"amount":0.000015,"unit":"outputToken"},{"amount":3e-7,"unit":"cachedInputToken"}]},{"model":"google-gemini-3-5-flash-lite","air":"google:gemini@3.5-flash-lite","name":"Gemini 3.5 Flash-Lite","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-07-21T00:00:00Z","weight":5,"headline":"Fast high-throughput Gemini model for low-latency agentic and document-heavy workloads","description":"Gemini 3.5 Flash-Lite is a lightweight multimodal model in Google's Gemini 3.5 family, built for high-throughput and latency-sensitive production workloads. It is well suited to agentic search, document processing, coding subtasks, routing, summarization, and other flows where fast response times and scale matter as much as model quality. It combines the Lite tier's efficiency profile with stronger long-context, coding, and agent-style performance than earlier Flash-Lite generations.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-5-flash-lite.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-5-flash-lite.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":3e-7,"unit":"inputToken"},{"amount":0.0000025,"unit":"outputToken"},{"amount":3e-8,"unit":"cachedInputToken"}]},{"model":"google-gemini-3-6-flash","air":"google:gemini@3.6-flash","name":"Gemini 3.6 Flash","creator":"google","endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-07-21T00:00:00Z","weight":5,"headline":"More efficient multimodal Flash model for coding, knowledge work, and agentic execution","description":"Gemini 3.6 Flash is an updated multimodal reasoning model in Google's Flash series for production agent workflows, coding, and knowledge-intensive tasks. It is designed to deliver stronger coding quality, better computer-use performance, and more efficient reasoning than the previous generation, with fewer unnecessary steps and lower token usage while preserving the Flash family's balance of speed, scale, and capability.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-6-flash.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-6-flash.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.0000015,"unit":"inputToken"},{"amount":0.0000075,"unit":"outputToken"},{"amount":1.5e-7,"unit":"cachedInputToken"}]},{"model":"bytedance-seedream-5-0-pro","air":"bytedance:seedream@5.0-pro","name":"Seedream 5.0 Pro","creator":"bytedance","status":"live","releasedAt":"2026-07-08T00:00:00Z","weight":5,"headline":"Flagship image model for controlled editing, multi-reference generation, and layer-aware visual workflows","description":"Seedream 5.0 Pro is ByteDance's flagship image generation and editing model for teams that need stronger control than prompt-only workflows provide. It supports text-to-image and reference-guided image editing with up to 10 input images, and is designed for precise local changes using coordinates, masks, bounding boxes, sketches, and material or color instructions. The model also supports multi-image fusion, advanced layer separation, and strong multilingual rendering, making it well suited to design, marketing, e-commerce, and storyboard-driven content production.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-5-0-pro.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-5-0-pro.png","pricingOverview":"Pricing starts at $0.04815 per image generated (1.5K output image that is smaller or equal to (<=) 2.36 MP) to $0.0963 per image (2K output image that is bigger than (>) 2.36 MP). First input image is free and any additional input image is $0.00321 per image.","pricingRates":[{"amount":0.04815,"unit":"output","label":"1.5K"},{"amount":0.0963,"unit":"output","label":"2K"},{"amount":0.00321,"unit":"inputImage","after":1}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.0481","amount":0.04815,"latencyMs":80196,"exampleId":"edwardian-station-vfx-clean-plate"},{"configuration":"text-to-image · 1536×1024","price":"$0.0481","amount":0.04815,"latencyMs":79834,"exampleId":"bioluminescent-algae-culture-kit-packaging"},{"configuration":"reference-to-image","price":"$0.0514","amount":0.05136,"latencyMs":131089,"exampleId":"museum-astrolabe-catalog-cleanup"},{"configuration":"reference-to-image","price":"$0.0514","amount":0.05136,"latencyMs":103756,"exampleId":"urban-beekeeper-linocut-profile-avatar"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Seedream 5.0 Pro for text-to-image, reference-guided editing, and multi-image fusion in commercial workflows: campaign posters, product variants, and editorial still-life.","date":"2026-07-08","url":"https://runware.ai/docs/models/bytedance-seedream-5-0-pro/guides/prompting"},{"slug":"multilingual-text-rendering","title":"Multilingual text rendering","description":"How to prompt Seedream 5.0 Pro for accurate in-image text across 15 native languages, from French posters with accented characters to Arabic, Thai, Korean, and Japanese scripts.","date":"2026-07-08","url":"https://runware.ai/docs/models/bytedance-seedream-5-0-pro/guides/multilingual-text-rendering"}]},{"model":"meta-muse-image","air":"meta:muse@image","name":"Muse Image","creator":"meta","status":"live","releasedAt":"2026-07-07T00:00:00Z","weight":5,"headline":"Agentic image generation and editing with multi-reference composition, clean text rendering, and precise local edits","description":"Muse Image is Meta's flagship image generation model from Meta Superintelligence Labs. It is built for prompt-faithful image creation, precision editing, and multi-reference composition, with strong text rendering and the ability to refine existing photos through localized markup-based edits. Meta positions it as an agentic image model that plans layouts, uses search and coding tools to improve accuracy, blends multiple visual references intelligently, and handles both creative generation and practical visual tasks such as infographics, QR codes, restorations, product-style mockups, and photobomber removal.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/meta-muse-image.jpg","ogImage":"https://assets.runware.ai/og/meta-muse-image.png","pricingRates":[{"amount":0.01,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 2048×1152","price":"$0.01","amount":0.01,"latencyMs":26011,"exampleId":"antikythera-exhibition-banner-expansion"},{"configuration":"reference-to-image · 1344×1792","price":"$0.01","amount":0.01,"latencyMs":18212,"exampleId":"cello-restoration-social-media-ad"},{"configuration":"text-to-image · 2048×1152","price":"$0.01","amount":0.01,"latencyMs":36396,"exampleId":"equine-ultrasound-campaign-hero"},{"configuration":"reference-to-image · 1344×1792","price":"$0.01","amount":0.01,"latencyMs":17992,"exampleId":"beekeeping-swarm-crew-shirt-graphic"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Muse Image: writing a brief it can plan a layout from, setting the thinking level, picking one of the eight size pairs, and working without a seed.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/prompting"},{"slug":"editing-images","title":"Editing images","description":"How to edit an image with Muse Image: passing one source image, targeting a region in words, removing objects, restoring old photos, and refining a result across passes.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/editing-images"},{"slug":"text-in-images","title":"Rendering exact text","description":"How to render exact, legible copy inside an image with Muse Image: quoting the literal string, keeping it short, directing the type, and landing multi-element layouts.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/text-in-images"},{"slug":"multi-reference-composition","title":"Composing with multiple reference images","description":"How to build one image out of up to ten reference images with Muse Image: giving each reference a job in the prompt, and letting the references set the output shape.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/multi-reference-composition"},{"slug":"grounded-generation","title":"Grounding images with web and image search","description":"How to ground Muse Image in real facts and real references with settings.webSearch and settings.imageSearch, and when to switch the lookups off.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/grounded-generation"},{"slug":"infographics-and-charts","title":"Building infographics and charts","description":"How to build data-accurate charts and infographics with Muse Image: what settings.shell does, how to write your figures into the prompt, and when to switch it off.","date":"2026-08-31","url":"https://runware.ai/docs/models/meta-muse-image/guides/infographics-and-charts"}]},{"model":"anthropic-claude-sonnet-5","air":"anthropic:claude@sonnet-5","name":"Claude Sonnet 5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"compatible-only","releasedAt":"2026-06-30T00:00:00Z","weight":5,"headline":"Balanced multimodal Claude model for everyday coding, agents, and production workloads","description":"Claude Sonnet 5 is Anthropic’s balanced model for daily production use, from coding and agent workflows to customer-facing assistants and high-volume automation. It reads text and images, calls tools, and supports extended thinking, at a fraction of the price of the Opus line. Its 1M-token context window and 128K-token maximum output cover long conversations, large documents, and multi-file code work.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-sonnet-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-sonnet-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken"},{"amount":0.00001,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"},{"amount":0.0000025,"unit":"cacheWriteToken"}]},{"model":"google-nano-banana-2-lite","air":"google:nano-banana@2-lite","name":"Nano Banana 2 Lite","creator":"google","zdr":true,"status":"live","releasedAt":"2026-06-30T00:00:00Z","weight":5,"headline":"Lighter Nano Banana 2 image model for faster generation and editing workflows","description":"Nano Banana 2 Lite is a lighter variant in Google's Nano Banana 2 image model family. It is positioned as a more efficient option for teams that want the same broad text-to-image and image-editing workflow shape as Nano Banana 2 but with faster turnaround and a smaller model footprint. It is best understood as a lower-latency, higher-throughput entry point into the Nano Banana 2 family rather than a separate creative direction.","capabilities":["io:text-to-image","io:image-to-image","io:video-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-nano-banana-2-lite.jpg","ogImage":"https://assets.runware.ai/og/google-nano-banana-2-lite.png","pricingBasis":"floor","pricingRates":[{"amount":2.5e-7,"unit":"inputToken"},{"amount":0.0000015,"unit":"outputToken","label":"Text output"},{"amount":0.00003,"unit":"outputToken","label":"Image output"},{"amount":0.033603,"unit":"output","label":"Example: 1K image output"},{"amount":0.000337,"unit":"output","label":"Example: 1K image input"}],"pricingExamples":[{"configuration":"text-to-image · 1376×768","price":"$0.0338","amount":0.0337506,"latencyMs":3992,"exampleId":"historic-dry-dock-lighting-visualization"},{"configuration":"text-to-image · 1376×768","price":"$0.0338","amount":0.0337695,"latencyMs":4305,"exampleId":"airport-wildlife-biologist-picture-book-spread"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0344","amount":0.0343875,"latencyMs":6508,"exampleId":"mineral-saltworks-square-wall-print"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0344","amount":0.0343986,"latencyMs":6939,"exampleId":"textile-conservation-journal-cover"}]},{"model":"zai-glm-5-2","air":"zai:glm@5.2","name":"GLM-5.2","creator":"zai","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-06-16T00:00:00Z","weight":5,"headline":"Flagship long-horizon coding LLM with usable 1M context and stronger end-to-end engineering execution","description":"GLM-5.2 is Z.ai's flagship language model for long-horizon coding, agentic engineering, and sustained multi-step execution. It is designed to keep large project context coherent over extended runs, with a 1M token context window, 128K max output, multiple thinking modes, function calling, structured output, context caching, streaming, and MCP support for tool-rich workflows.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/zai-glm-5-2.jpg","ogImage":"https://assets.runware.ai/og/zai-glm-5-2.png","contextWindow":1048576,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":8e-7,"unit":"inputToken"},{"amount":0.00000255,"unit":"outputToken"},{"amount":1.6e-7,"unit":"cachedInputToken"}]},{"model":"anthropic-claude-fable-5","air":"anthropic:claude@fable-5","name":"Claude Fable 5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"live","releasedAt":"2026-06-09T00:00:00Z","weight":5,"headline":"Frontier multimodal agentic model for long-horizon coding, research, and complex professional work","description":"Claude Fable 5 is Anthropic's new top-capability generally available Claude model. It is built for long-running coding, agentic execution, multimodal reasoning, research, and high-stakes professional workflows, with stronger long-horizon performance than earlier Opus models, state-of-the-art vision, and a 1M-token context window. Safety classifiers can route some sensitive requests to Claude Opus 4.8 instead.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-fable-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-fable-5.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.00001,"unit":"inputToken"},{"amount":0.00005,"unit":"outputToken"},{"amount":0.000001,"unit":"cachedInputToken"},{"amount":0.0000125,"unit":"cacheWriteToken"}],"pricingExamples":[{"configuration":"split-thinking","price":"$0.0347","amount":0.03466,"latencyMs":10709,"exampleId":"commercial-solar-bid-recommendation"},{"configuration":"text-generation","price":"$0.119","amount":0.11926,"latencyMs":35389,"exampleId":"library-capital-campaign-case-statement"},{"configuration":"structured-output","price":"$0.14","amount":0.14022,"latencyMs":33445,"exampleId":"mobile-prenatal-clinic-siting-brief"},{"configuration":"reference-to-text","price":"$0.271","amount":0.2705,"latencyMs":84501,"exampleId":"freight-damage-claims-assessment"}]},{"model":"ideogram-4-0","air":"ideogram:4@0","name":"Ideogram 4.0","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":5,"headline":"Open-weight design-focused text-to-image model with strong typography, layout control, transparent backgrounds, and 2K output","description":"Ideogram 4.0 is Ideogram's most capable text-to-image model for design-heavy image generation. It is built for frontier text rendering across languages, structured prompt control through natural language or JSON, bounding-box layout control, transparent background generation, and high-fidelity 2K output. It is well suited to posters, branded graphics, packaging, product visuals, typography-led compositions, and other workflows where design precision matters as much as visual quality.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-4-0.jpg","ogImage":"https://assets.runware.ai/og/ideogram-4-0.png","pricingOverview":"Pricing is $0.03 per image for Turbo, $0.06 per image for Default, and $0.10 per image for Quality\n","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.1,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"text-to-image · 1440p","price":"$0.06","amount":0.06,"latencyMs":25159,"exampleId":"municipal-sewer-crew-sticker-pack"},{"configuration":"text-to-image · 1440p","price":"$0.06","amount":0.06,"latencyMs":18303,"exampleId":"thermal-battery-video-thumbnail"},{"configuration":"text-to-image · 1440p","price":"$0.06","amount":0.06,"latencyMs":18412,"exampleId":"alpine-survival-game-key-art"},{"configuration":"structured-prompt · 1792×2240","price":"$0.10","amount":0.1,"latencyMs":17874,"exampleId":"community-radio-host-campaign-portrait"}],"guides":[{"slug":"structured-prompts","title":"Structured prompts","description":"How Ideogram 4.0's two prompting modes work: natural language with Magic Prompt expansion for quick exploration, and the full trained JSON schema for explicit per-element control.","date":"2026-06-03","url":"https://runware.ai/docs/models/ideogram-4-0/guides/structured-prompts"},{"slug":"text-and-design","title":"Text and design output","description":"How to use Ideogram 4.0 for typography-heavy design where the text has to be readable and exactly right, the layout has to land, and the palette has to lock to brand.","date":"2026-06-03","url":"https://runware.ai/docs/models/ideogram-4-0/guides/text-and-design"}]},{"model":"fish-audio-s2-1-pro","air":"fishaudio:s2.1@pro","name":"Fish Audio S2.1 Pro","creator":"fish-audio","status":"live","releasedAt":"2026-06-01T00:00:00Z","weight":5,"headline":"Flagship multilingual text-to-speech with natural language voice control and realtime streaming","description":"Fish Audio S2.1 Pro is a flagship text-to-speech model built for highly expressive, low-latency speech generation. It supports natural-language bracket cues for emotion and delivery control, multi-speaker dialogue in a single generation, 80+ languages with automatic language detection, and realtime streaming with very fast time to first audio.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/fish-audio-s2-1-pro.jpg","ogImage":"https://assets.runware.ai/og/fish-audio-s2-1-pro.png","pricingOverview":"Pricing is based on the size of input text, measured in millions of UTF-8 bytes. ","pricingRates":[{"amount":0.000015,"unit":"utf8Byte"}],"pricingExamples":[{"configuration":"multi-speaker-dialogue","price":"$0.00356","amount":0.003555,"latencyMs":3688,"exampleId":"midnight-cinema-channel-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.0168","amount":0.016785,"latencyMs":13740,"exampleId":"reservoir-science-podcast-cold-open"},{"configuration":"voice-cloning","price":"$0.0171","amount":0.017145,"latencyMs":18082,"exampleId":"bronze-bell-inspection-course-narration"},{"configuration":"text-to-audio","price":"$0.0227","amount":0.02268,"latencyMs":20299,"exampleId":"forensic-entomology-training-narration"}],"guides":[{"slug":"emotion-and-expression","title":"Emotion and expression control","description":"How to control vocal delivery in Fish Audio S2-Pro with bracket tags. The tag system steers emotion, expression, paralanguage, and phoneme-level pronunciation in one inline syntax.","date":"2026-06-01","url":"https://runware.ai/docs/models/fish-audio-s2-1-pro/guides/emotion-and-expression"},{"slug":"multi-speaker-dialogue","title":"Multi-speaker dialogue","description":"How to generate two-speaker dialogue audio in a single request to Fish Audio S2-Pro using inline speaker tags. One call, two voices, full per-speaker emotion control.","date":"2026-06-01","url":"https://runware.ai/docs/models/fish-audio-s2-1-pro/guides/multi-speaker-dialogue"}]},{"model":"minimax-m3","air":"minimax:m3@0","name":"MiniMax M3","creator":"minimax","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-06-01T00:00:00Z","weight":5,"headline":"Frontier multimodal agentic model with 1M context for coding, long-horizon tasks, and visual understanding","description":"MiniMax M3 is MiniMax's new flagship open-weight model for coding, agentic execution, and multimodal reasoning. It is built around the MiniMax Sparse Attention architecture, supports up to 1 million tokens of context, accepts image and video input in addition to text, and is designed for long-running software, research, browsing, and desktop-operation workflows that need strong tool use and sustained multi-step performance.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-m3.jpg","ogImage":"https://assets.runware.ai/og/minimax-m3.png","contextWindow":1000000,"maxOutputTokens":512000,"pricingBasis":"token","pricingRates":[{"amount":3e-7,"unit":"inputToken","label":"≤ 512K context"},{"amount":6e-7,"unit":"inputToken","label":"> 512K context"},{"amount":0.0000012,"unit":"outputToken","label":"≤ 512K context"},{"amount":0.0000024,"unit":"outputToken","label":"> 512K context"},{"amount":6e-8,"unit":"cachedInputToken","label":"≤ 512K context"},{"amount":1.2e-7,"unit":"cachedInputToken","label":"> 512K context"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000136","amount":0.000136,"latencyMs":1230,"exampleId":"forced-utc-time-function-call"},{"configuration":"reference-to-text","price":"$0.00724","amount":0.007244,"latencyMs":59710,"exampleId":"warehouse-rack-incident-investigation-brief"},{"configuration":"text-generation","price":"$0.0107","amount":0.010739,"latencyMs":71581,"exampleId":"boutique-hotel-investment-committee-memo"},{"configuration":"text-generation","price":"$0.012","amount":0.01203,"latencyMs":86425,"exampleId":"hospital-ransomware-tabletop-exercise-package"}]},{"model":"xai-grok-imagine-video-1-5","air":"xai:grok-imagine@video-1.5","name":"Grok Imagine Video 1.5","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-05-30T00:00:00Z","weight":5,"headline":"Higher-tier Grok image-to-video generation from a single starting frame with longer durations and stronger output quality","description":"Grok Imagine Video 1.5 is xAI's newer image-to-video model. It is positioned above the earlier Grok Imagine Video release with higher per-second pricing, supports durations up to 15 seconds, and generates 480p or 720p video from a single still-image starting frame for cinematic clips, animated visuals, and prompt-guided short-form video creation.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/xai-grok-imagine-video-1-5-preview.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-imagine-video-1-5-preview.png","pricingOverview":"Pricing starts at $0.08 per second of output at 480p, $0.14 per second of output at 720p, and $0.25 per second of output at 1080p. Reference image for I2V costs $0.01 per image on top.\n","pricingRates":[{"amount":0.08,"unit":"durationSecond","label":"480p"},{"amount":0.14,"unit":"durationSecond","label":"720p"},{"amount":0.25,"unit":"durationSecond","label":"1080p"},{"amount":0.01,"unit":"inputImage"}],"pricingExamples":[{"configuration":"reference-to-video · 720p · 5s","price":"$0.73","amount":0.73,"latencyMs":43388,"exampleId":"avalanche-forecast-center-logo-loop"},{"configuration":"first-frame · 720p · 7s","price":"$0.99","amount":0.99,"latencyMs":41676,"exampleId":"mobile-knife-sharpening-social-reel"},{"configuration":"multi-speaker-dialogue · 720p · 10s","price":"$1.40","amount":1.4,"latencyMs":55451,"exampleId":"vertical-oyster-hatchery-b-roll"},{"configuration":"text-to-video · 1080p · 6s","price":"$1.50","amount":1.5,"latencyMs":222127,"exampleId":"raku-workshop-social-video-ad"}]},{"model":"anthropic-claude-opus-4-8","air":"anthropic:claude@opus-4.8","name":"Claude Opus 4.8","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"live","releasedAt":"2026-05-28T00:00:00Z","weight":5,"headline":"Frontier multimodal reasoning model for advanced coding, agent systems, and long-context knowledge work","description":"Claude Opus 4.8 is Anthropic's highest-capability Claude model. It is built for demanding coding, agent orchestration, multimodal reasoning, and professional workflows that need strong instruction following, adaptive and extended thinking, high-resolution vision, and a 1M-token context window.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-opus-4-8.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-opus-4-8.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken"},{"amount":0.000025,"unit":"outputToken"},{"amount":5e-7,"unit":"cachedInputToken"},{"amount":0.00000625,"unit":"cacheWriteToken"}],"pricingExamples":[{"configuration":"reference-to-text","price":"$0.0849","amount":0.0849,"latencyMs":51604,"exampleId":"marathon-sponsorship-visibility-audit"},{"configuration":"reference-to-text","price":"$0.0851","amount":0.08514,"latencyMs":47996,"exampleId":"electrical-panel-commissioning-punch-list"},{"configuration":"text-generation","price":"$0.158","amount":0.158425,"latencyMs":63234,"exampleId":"food-recall-communications-package"},{"configuration":"text-generation","price":"$0.165","amount":0.16535,"latencyMs":102329,"exampleId":"prestige-feature-script-coverage"}]},{"model":"google-gemini-3-5-flash","air":"google:gemini@3.5-flash","name":"Gemini 3.5 Flash","creator":"google","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-05-19T00:00:00Z","weight":5,"headline":"Frontier multimodal reasoning model for agentic and coding workflows","description":"Gemini 3.5 Flash is Google’s most intelligent Flash-series multimodal model for sustained frontier performance on agentic and coding tasks. It accepts text, images, video, audio, and PDFs, and is designed for long-horizon workflows, sub-agent orchestration, complex coding loops, multimodal understanding, and high-speed reasoning at production scale.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-5-flash.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-5-flash.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.0000015,"unit":"inputToken"},{"amount":0.000009,"unit":"outputToken"},{"amount":1.5e-7,"unit":"cachedInputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.00106","amount":0.001059,"latencyMs":13784,"exampleId":"production-service-health-lookup"},{"configuration":"audio-driven","price":"$0.0148","amount":0.014816,"latencyMs":11729,"exampleId":"wind-turbine-service-memo-summary"},{"configuration":"text-generation","price":"$0.033","amount":0.033038,"latencyMs":24369,"exampleId":"saas-renewal-negotiation-brief"},{"configuration":"structured-output","price":"$0.0359","amount":0.035916,"latencyMs":25056,"exampleId":"commercial-lease-risk-analysis"},{"configuration":"reference-to-text","price":"$0.0365","amount":0.036509,"latencyMs":33230,"exampleId":"retail-planogram-compliance-report"},{"configuration":"search-grounded","price":"$0.307","amount":0.307268,"latencyMs":62232,"exampleId":"savannah-container-terminal-status-brief"}]},{"model":"inworld-tts-2","air":"inworld:tts@2","name":"Inworld Realtime TTS-2","creator":"inworld","status":"live","releasedAt":"2026-05-05T00:00:00Z","weight":5,"headline":"Conversational text-to-speech with realtime voice direction and audio-aware delivery","description":"Inworld Realtime TTS-2 is a conversational text-to-speech model built for realtime voice interaction rather than static narration. It supports free-form voice direction, carries tone and pacing forward from prior audio in a session, preserves one voice identity across 100+ languages, and is designed for expressive, low-latency speech in assistants, characters, support agents, and interactive products.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/inworld-tts-2.jpg","ogImage":"https://assets.runware.ai/og/inworld-tts-2.png","pricingRates":[{"amount":0.000035,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.00154","amount":0.00154,"latencyMs":2307,"exampleId":"immersive-greenhouse-warning-cue"},{"configuration":"text-to-audio","price":"$0.00263","amount":0.002625,"latencyMs":1752,"exampleId":"roller-derby-livestream-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.00507","amount":0.005075,"latencyMs":3721,"exampleId":"polish-numbers-station-voice-loop"},{"configuration":"text-to-audio","price":"$0.00577","amount":0.005775,"latencyMs":3711,"exampleId":"italian-heritage-podcast-intro"}],"guides":[{"slug":"llm-system-prompts","title":"Formatting LLM output for speech","description":"How to write LLM system prompts that produce text TTS-2 can synthesize naturally, with normalization, filler words, and emphasis cues handled before the audio call.","date":"2026-05-20","url":"https://runware.ai/docs/models/inworld-tts-2/guides/llm-system-prompts"},{"slug":"voice-direction","title":"Controlling voice delivery with steering tags","description":"How to use natural-language steering tags to control emotion, pacing, volume, and vocal style in TTS-2 speech output.","date":"2026-05-22","url":"https://runware.ai/docs/models/inworld-tts-2/guides/voice-direction"}]},{"model":"deepseek-v4-pro","air":"deepseek:v4@pro","name":"DeepSeek-V4-Pro-0813","creator":"deepseek","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-04-24T00:00:00Z","weight":5,"headline":"High-capability frontier LLM with 1M context, stronger agent performance, and dual thinking modes","description":"DeepSeek-V4-Pro is DeepSeek's flagship V4 language model for coding, reasoning, and agent workflows that need stronger overall capability than the Flash variant. It supports both thinking and non-thinking modes, a 1M token context window, up to 384K output tokens, tool calls, JSON output, and long-context operation for demanding software, research, and structured professional workloads.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/deepseek-v4-pro.jpg","ogImage":"https://assets.runware.ai/og/deepseek-v4-pro.png","contextWindow":1048576,"maxOutputTokens":384000,"pricingBasis":"token","pricingRates":[{"amount":9.61e-7,"unit":"inputToken"},{"amount":0.000001922,"unit":"outputToken"},{"amount":7.9e-8,"unit":"cachedInputToken"}]},{"model":"openai-gpt-5-5","air":"openai:gpt@5.5","name":"GPT-5.5","creator":"openai","endpoints":["chat-completions","responses"],"status":"live","releasedAt":"2026-04-23T00:00:00Z","weight":5,"headline":"Frontier reasoning LLM for complex coding, long-context work, and tool-using professional tasks","description":"GPT-5.5 is OpenAI's newest frontier model for complex professional work, with strong performance in coding, reasoning, and tool-using workflows. It supports a 1,050,000 token context window, 128,000 max output tokens, configurable reasoning effort, image input, and a broad tool stack including web search, file search, code interpreter, hosted shell, apply patch, skills, MCP, tool search, and computer use.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-5.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-5.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken"},{"amount":0.00003,"unit":"outputToken"},{"amount":5e-7,"unit":"cachedInputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.00128","amount":0.001275,"latencyMs":2475,"exampleId":"location-shoot-weather-tool-handoff"},{"configuration":"reference-to-text","price":"$0.0699","amount":0.06992,"latencyMs":33795,"exampleId":"vineyard-disease-scouting-report"},{"configuration":"structured-output","price":"$0.134","amount":0.13368,"latencyMs":43902,"exampleId":"fleet-telematics-bid-evaluation"},{"configuration":"text-generation","price":"$0.808","amount":0.80791,"latencyMs":335148,"exampleId":"biotech-licensing-decision-memo"}]},{"model":"openai-gpt-image-2","air":"openai:gpt-image@2","name":"GPT Image 2","creator":"openai","status":"deprecated","replacedBy":"openai-gpt-image-2-6-sunburst","deactivatesAt":"2027-02-10T00:00:00Z","releasedAt":"2026-04-21T00:00:00Z","weight":5,"headline":"Image generation and editing model with strong prompt fidelity, text rendering, and layout-aware control","description":"GPT Image 2 is a general-purpose GPT Image family model for text-to-image generation and image editing. Its strengths include strong prompt adherence, readable embedded text, detailed edits, photorealistic rendering, and structured visual outputs such as posters, packaging, product comps, diagrams, and other layout-sensitive images.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-2.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-2.png","pricingOverview":"Pricing is based per 1M Token for image at $8.00 (input), $2.00 (cached input), $30.00 (output). For text at $5.00 (input), $1.25 (cached input), $10.00 (output). Below are estimation examples based on quality which changes significantly the price.","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.00003,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"transparent-background","price":"$0.0221","amount":0.022145,"latencyMs":34999,"exampleId":"rail-signal-training-character-portrait"},{"configuration":"inpaint","price":"$0.0223","amount":0.022282,"latencyMs":22810,"exampleId":"municipal-water-tower-lighting-visualization"},{"configuration":"text-to-image","price":"$0.0426","amount":0.042605,"latencyMs":44364,"exampleId":"wadden-sea-walking-campaign-hero"},{"configuration":"reference-to-image","price":"$0.0675","amount":0.067516,"latencyMs":42698,"exampleId":"blackstart-grid-strategy-key-art"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for GPT Image 2 across the cases the model handles unusually well: photorealism, accurate text, world-knowledge composition, and multi-image editing workflows.","date":"2026-05-21","url":"https://runware.ai/docs/models/openai-gpt-image-2/guides/prompting"}]},{"model":"anthropic-claude-opus-4-7","air":"anthropic:claude@opus-4.7","name":"Claude Opus 4.7","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"live","releasedAt":"2026-04-16T00:00:00Z","weight":5,"headline":"Frontier multimodal reasoning model for advanced coding, long-running agents, and complex knowledge work","description":"Claude Opus 4.7 is Anthropic's highest-capability generally available Claude model. It is designed for demanding coding, agent orchestration, multimodal reasoning, and high-stakes professional workflows, with stronger instruction following, better high-resolution vision, adaptive thinking, and a 1M-token context window.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-opus-4-7.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-opus-4-7.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken"},{"amount":0.000025,"unit":"outputToken"},{"amount":5e-7,"unit":"cachedInputToken"},{"amount":0.00000625,"unit":"cacheWriteToken"}],"pricingExamples":[{"configuration":"reference-to-text","price":"$0.0609","amount":0.06087,"latencyMs":35866,"exampleId":"beverage-shelf-compliance-audit"},{"configuration":"reference-to-text","price":"$0.078","amount":0.078035,"latencyMs":47830,"exampleId":"textile-conservation-intake-report"},{"configuration":"text-generation","price":"$0.438","amount":0.438005,"latencyMs":229785,"exampleId":"enterprise-identity-consolidation-execution-plan"},{"configuration":"text-generation","price":"$0.51","amount":0.509515,"latencyMs":220310,"exampleId":"payment-webhook-reliability-remediation-plan"}]},{"model":"google-gemini-3-1-flash-tts","air":"google:gemini@3.1-flash-tts","name":"Gemini 3.1 Flash TTS","creator":"google","zdr":true,"status":"deprecated","deactivatesAt":"2026-11-17T00:00:00Z","releasedAt":"2026-04-15T00:00:00Z","weight":5,"headline":"Expressive text-to-speech with audio tags, multi-speaker dialogue, and 70+ languages","description":"Gemini 3.1 Flash TTS is a text-to-speech model for expressive spoken audio generation from text. It supports granular control over delivery through audio tags, native multi-speaker dialogue, and speech generation across 70+ languages, making it suitable for narration, conversational voice apps, podcasts, audiobooks, and other production-oriented voice workflows.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-1-flash-tts.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-1-flash-tts.png","pricingBasis":"token","pricingRates":[{"amount":0.000001,"unit":"inputToken"},{"amount":0.00002,"unit":"outputToken"}],"pricingExamples":[{"configuration":"multi-speaker-dialogue","price":"$0.00621","amount":0.006207,"latencyMs":5757,"exampleId":"repair-cooperative-podcast-vocal-stinger"},{"configuration":"multi-speaker-dialogue","price":"$0.00792","amount":0.00792,"latencyMs":6695,"exampleId":"volcano-observatory-podcast-stinger"},{"configuration":"text-to-audio","price":"$0.0251","amount":0.025115,"latencyMs":31150,"exampleId":"oyster-farm-insurance-explainer"},{"configuration":"text-to-audio","price":"$0.0278","amount":0.027822,"latencyMs":27511,"exampleId":"ice-rink-heat-recovery-explainer"}]},{"model":"zai-glm-5-1","air":"zai:glm@5.1","name":"GLM-5.1","creator":"zai","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-04-07T00:00:00Z","weight":5,"headline":"Flagship agentic coding model with 200K context, deep thinking, and long-horizon task execution","description":"GLM-5.1 is Z.ai’s flagship language model for agentic engineering, coding, reasoning, and tool-driven workflows. It supports a 200K token context window with up to 128K output tokens, deep thinking, function calling, structured output, and streaming tool calls, and is designed to stay effective over long multi-step sessions rather than only short-horizon tasks.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/zai-glm-5-1.jpg","ogImage":"https://assets.runware.ai/og/zai-glm-5-1.png","contextWindow":204800,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":0.0000014,"unit":"inputToken"},{"amount":0.0000044,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000342","amount":0.0003416,"latencyMs":5386,"exampleId":"field-operations-weather-tool-call"},{"configuration":"tool-calling","price":"$0.000409","amount":0.0004094,"latencyMs":4767,"exampleId":"wedding-weekend-weather-tool-call"},{"configuration":"text-generation","price":"$0.0019","amount":0.0018994,"latencyMs":15439,"exampleId":"board-cyber-incident-briefing"},{"configuration":"text-generation","price":"$0.00257","amount":0.0025744,"latencyMs":21879,"exampleId":"boutique-hotel-loan-credit-memo"}]},{"model":"bytedance-seedance-2-0","air":"bytedance:seedance@2.0","name":"Seedance 2.0","creator":"bytedance","status":"live","releasedAt":"2026-04-03T00:00:00Z","weight":5,"headline":"Unified multimodal audio-video generation with multi-reference input and physics-aware motion","description":"Seedance 2.0 is a unified multimodal audio-video generation model from ByteDance that accepts text, image, audio, and video inputs in combination, supporting up to 9 images, 3 video clips, and 3 audio clips as reference. It generates multi-shot videos up to 15 seconds with dual-channel synchronized audio including dialogue, ambient sound, and effects. It features physics-aware motion, improved controllability for video extension and editing, and strong instruction following for complex scene composition.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-2-0.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-2-0.png","pricingOverview":"Text/Image to Video, 480p: $0.07/sec\nText/Image to Video, 720p: $0.16/sec\nText/Image to Video, 1080p: $0.40/sec\nText/Image to Video, 4K: $0.836/sec\nVideo-to-Video, 480p: from $0.09/sec\nVideo-to-Video, 720p: from $0.20/sec\nVideo-to-Video, 1080p: from $0.48/sec\nVideo-to-Video, 4K: from $1.050/sec\n\nVideo-to-Video rates assume reference video matches output duration ","pricingRates":[{"amount":0.07,"unit":"durationSecond","label":"480p"},{"amount":0.16,"unit":"durationSecond","label":"720p"},{"amount":0.4,"unit":"durationSecond","label":"1080p"},{"amount":0.8559,"unit":"durationSecond","label":"4K"},{"amount":0.21942,"unit":"durationSecond","label":"720p · video to video"},{"amount":0.53663,"unit":"durationSecond","label":"1080p · video to video"},{"amount":1.15911,"unit":"durationSecond","label":"4K · video to video"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 6s · audio","price":"$0.953","amount":0.95265,"latencyMs":123407,"exampleId":"tattoo-machine-colorway-reveal-film"},{"configuration":"first-last-frame · 720p · 8s","price":"$1.27","amount":1.26801,"latencyMs":137222,"exampleId":"secretary-bird-habitat-composite-film"},{"configuration":"first-frame · 1080p · 5s · audio","price":"$2.03","amount":2.0337075,"latencyMs":213917,"exampleId":"art-forgery-drama-title-sequence"},{"configuration":"reference-to-video · 1080p · 10s · audio","price":"$4.14","amount":4.141125,"latencyMs":324541,"exampleId":"atp-sanitation-meter-demo"}]},{"model":"openai-gpt-5-4","air":"openai:gpt@5.4","name":"GPT-5.4","creator":"openai","endpoints":["chat-completions","responses"],"status":"live","releasedAt":"2026-03-05T00:00:00Z","weight":5,"headline":"Flagship reasoning LLM with 1M context, native computer use, and high factual accuracy","description":"GPT-5.4 is OpenAI's flagship large language model, featuring a 1 million token context window, native computer use, and a 33% reduction in factual errors over GPT-5.2. It integrates coding capabilities from GPT-5.3-Codex, is 47% more token-efficient, and supports configurable reasoning effort for complex professional tasks.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-4.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-4.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.0000025,"unit":"inputToken"},{"amount":0.000015,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000543","amount":0.000543,"latencyMs":1183,"exampleId":"production-deployment-utc-attestation"},{"configuration":"reference-to-text","price":"$0.0278","amount":0.02782,"latencyMs":23427,"exampleId":"residential-solar-compliance-memo"},{"configuration":"text-generation","price":"$0.286","amount":0.286055,"latencyMs":153551,"exampleId":"credit-union-annual-report-copy"},{"configuration":"text-generation","price":"$0.392","amount":0.39209,"latencyMs":263687,"exampleId":"hospital-ransomware-tabletop-exercise-package"}]},{"model":"google-nano-banana-2","air":"google:4@3","name":"Nano Banana 2","creator":"google","zdr":true,"status":"live","releasedAt":"2026-02-26T00:00:00Z","weight":5,"headline":"Gemini 3.1 Flash Image fast high quality AI image generation and editing","description":"Nano Banana 2 (officially known as Gemini 3.1 Flash Image) is Google’s upgraded AI image generation and editing model that brings advanced visual creation capabilities to a broad audience. It generates detailed, expressive images from text and image prompts with sharp details, richer lighting, and improved adherence to complex instructions. Nano Banana 2 also supports multi-object and multi-character consistency, accurate text rendering within images, and flexible resolution control up to 4K. It is now integrated across Google’s AI platforms including the Gemini app, Search AI Mode, and other Gemini-powered services.","capabilities":["io:text-to-image","io:image-to-image","io:video-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-nano-banana-2.jpg","ogImage":"https://assets.runware.ai/og/google-nano-banana-2.png","pricingOverview":"Pricing starts at $0.04657 for 512x512. For every input image used, it's an additional $0.00028. When using grounded search, $0.14 will be added on top.","pricingBasis":"floor","pricingRates":[{"amount":0.04657,"unit":"output","label":"512x512"},{"amount":0.06895,"unit":"output","label":"1K"},{"amount":0.10255,"unit":"output","label":"2K"},{"amount":0.15295,"unit":"output","label":"4K"},{"amount":0.00028,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1376×768","price":"$0.0689","amount":0.06895,"latencyMs":25151,"exampleId":"underground-salt-cathedral-campaign-hero"},{"configuration":"text-to-image · 1376×768","price":"$0.0689","amount":0.06895,"latencyMs":31867,"exampleId":"winter-silo-lighting-visualization"},{"configuration":"reference-to-image · 2K","price":"$0.103","amount":0.10283,"latencyMs":26505,"exampleId":"orchard-robot-case-study-hero"},{"configuration":"reference-to-image · 2K","price":"$0.103","amount":0.10283,"latencyMs":53809,"exampleId":"subsea-cable-repair-explainer-thumbnail"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write prompts for Nano Banana 2: detailed scene descriptions, structured layering, legible text rendering, and thinking-level control.","date":"2026-06-15","url":"https://runware.ai/docs/models/google-nano-banana-2/guides/prompting"},{"slug":"multi-image-composition","title":"Combining multiple images into one composition","description":"How to merge several reference images, a product, a subject, a backdrop, or a style, into a single coherent image with Nano Banana 2.","date":"2026-06-15","url":"https://runware.ai/docs/models/google-nano-banana-2/guides/multi-image-composition"},{"slug":"character-consistency","title":"Keeping characters and products consistent","description":"How to use Nano Banana 2 reference images to keep the same character or product identical across new scenes and styles.","date":"2026-06-15","url":"https://runware.ai/docs/models/google-nano-banana-2/guides/character-consistency"},{"slug":"grounded-generation","title":"Grounded generation with web and image search","description":"How to generate images from real-time information with Nano Banana 2 using web and image search grounding via providerSettings.google.","date":"2026-06-15","url":"https://runware.ai/docs/models/google-nano-banana-2/guides/grounded-generation"}]},{"model":"anthropic-claude-sonnet-4-6","air":"anthropic:claude@sonnet-4.6","name":"Claude Sonnet 4.6","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":5,"headline":"Versatile multimodal language model for coding, agents, computer use, and production reasoning","description":"Claude Sonnet 4.6 is Anthropic's most capable Sonnet model, built for daily production use across coding, agent workflows, long-context reasoning, computer use, and professional knowledge work. It supports adaptive and extended thinking, strong instruction following, high-volume automation, and a 1M-token context window in beta.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-sonnet-4-6.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-sonnet-4-6.png","contextWindow":1000000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000003,"unit":"inputToken"},{"amount":0.000015,"unit":"outputToken"},{"amount":3e-7,"unit":"cachedInputToken"},{"amount":0.00000375,"unit":"cacheWriteToken"}],"pricingExamples":[{"configuration":"reference-to-text","price":"$0.0579","amount":0.057879,"latencyMs":90829,"exampleId":"preliminary-painting-condition-report"},{"configuration":"reference-to-text","price":"$0.0743","amount":0.074253,"latencyMs":107836,"exampleId":"restaurant-entrance-accessibility-audit"},{"configuration":"text-generation","price":"$0.246","amount":0.245898,"latencyMs":364798,"exampleId":"neighborhood-bookstore-turnaround-plan"},{"configuration":"text-generation","price":"$0.324","amount":0.324264,"latencyMs":395528,"exampleId":"solar-project-investment-committee-memo"}]},{"model":"openai-gpt-image-1-5","air":"openai:4@1","name":"GPT Image 1.5","creator":"openai","status":"deprecated","replacedBy":"openai-gpt-image-2","deactivatesAt":"2026-12-01T00:00:00Z","releasedAt":"2025-12-16T00:00:00Z","weight":5,"headline":"GPT Image 1.5 flagship image model with faster generation and enhanced editing","description":"GPT Image 1.5 is OpenAI’s newest flagship image model powering the latest ChatGPT Images. It delivers significantly faster image generation with stronger instruction following, more precise edits that preserve original details, more believable transformations, and improved rendering of dense or small text. It is suited for practical creative workflows, detailed design tasks, and production use cases.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-1-5.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-1-5.png","pricingOverview":"Input costs: ~$0.005/1k tokens (text) & ~$0.008/1k tokens (image). Added on top of output costs.","pricingBasis":"token","pricingRates":[{"amount":0.000005,"unit":"inputToken","label":"Text input"},{"amount":0.000008,"unit":"inputToken","label":"Image input"},{"amount":0.000032,"unit":"outputToken","label":"Image output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0575","amount":0.057549,"latencyMs":27740,"exampleId":"ferry-crew-archive-restoration"},{"configuration":"text-to-image","price":"$0.135","amount":0.134945,"latencyMs":37207,"exampleId":"quilted-lineworker-archive-portrait"},{"configuration":"text-to-image","price":"$0.136","amount":0.135535,"latencyMs":33318,"exampleId":"marine-emergency-signal-kit-packshot"},{"configuration":"inpaint","price":"$0.204","amount":0.203799,"latencyMs":47580,"exampleId":"wildlife-hospital-cut-paper-artwork"}]},{"model":"google-nano-banana-pro","air":"google:4@2","name":"Nano Banana Pro","creator":"google","zdr":true,"status":"live","releasedAt":"2025-11-20T00:00:00Z","weight":5,"headline":"Nano Banana Pro image preview for precise visual control","description":"Nano Banana Pro (also known as Nano Banana 2) is a Gemini 3 Pro Image Preview model for controlled visual creation. It improves reasoning over lighting and camera angle. It supports high resolution output and multi image blending for production ready design workflows and creative tools.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-nano-banana-pro.jpg","ogImage":"https://assets.runware.ai/og/google-nano-banana-pro.png","pricingOverview":"From $0.138/image","pricingBasis":"floor","pricingRates":[{"amount":0.138,"unit":"output","label":"1K"},{"amount":0.138,"unit":"output","label":"2K"},{"amount":0.244,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"text-to-image · 1376×768","price":"$0.138","amount":0.138,"latencyMs":31142,"exampleId":"sail-assisted-freighter-woodblock-feature"},{"configuration":"text-to-image · 1376×768","price":"$0.138","amount":0.138,"latencyMs":47325,"exampleId":"semiquincentennial-numismatics-magazine-cover"},{"configuration":"reference-to-image · 2K","price":"$0.14","amount":0.1402,"latencyMs":47778,"exampleId":"master-bell-tuner-headshot"},{"configuration":"reference-to-image · 2K","price":"$0.14","amount":0.1402,"latencyMs":47145,"exampleId":"freshwater-mussel-classroom-science-poster"}]},{"model":"bytedance-seedance-2-5-flash","air":"bytedance:seedance@2.5-flash","name":"Seedance 2.5 Flash","creator":"bytedance","status":"live","releasedAt":"2026-11-15T00:00:00Z","weight":4,"headline":"Lower-cost Seedance 2.5 video generation with 30-second clips, multimodal references, and video editing","description":"Seedance 2.5 Flash is a lighter, lower-cost variant of ByteDance's Seedance 2.5 video model. It keeps the same multimodal workflow, generating from a prompt, first and last frames, or up to 10 reference images, 5 reference videos and 5 reference audio clips, and it edits or extends an existing clip, producing 4 to 30 second videos with synchronized audio at 480p or 720p. It suits high-volume iteration, previews, and social content where the higher Seedance 2.5 resolutions are not needed.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-2-5-flash.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-2-5-flash.png","pricingRates":[{"amount":0.06781,"unit":"durationSecond","label":"Text/Image to Video · 480p"},{"amount":0.15246,"unit":"durationSecond","label":"Text/Image to Video · 720p"},{"amount":0.08331,"unit":"durationSecond","label":"Video-to-Video · 480p"},{"amount":0.18731,"unit":"durationSecond","label":"Video-to-Video · 720p"}],"pricingExamples":[{"configuration":"first-last-frame · 480p · 7s","price":"$0.474","amount":0.473564,"latencyMs":74249,"exampleId":"alpine-seed-bank-seasonal-visualization"},{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.762","amount":0.7623,"latencyMs":79079,"exampleId":"automatic-flood-barrier-product-demo"},{"configuration":"first-frame · 720p · 6s","price":"$0.913","amount":0.9135,"latencyMs":83037,"exampleId":"seed-vault-film-stock-transition"},{"configuration":"video-edit · 720p · autos","price":"$1.09","amount":1.08747,"latencyMs":81523,"exampleId":"predawn-vineyard-frost-patrol"},{"configuration":"video-extend · 720p · 6s","price":"$1.11","amount":1.11456,"latencyMs":76292,"exampleId":"archival-textile-conservation-b-roll"},{"configuration":"reference-to-video · 720p · 8s","price":"$1.30","amount":1.30419,"latencyMs":108193,"exampleId":"groundwater-documentary-teaser-sequence"}]},{"model":"xai-grok-imagine-video-1-5-lite","air":"xai:grok-imagine@video-1.5-lite","name":"Grok Imagine Video 1.5 Lite","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-10-02T00:00:00Z","weight":4,"headline":"Lower-cost Grok video generation from text or a single starting frame, with native audio and clips up to 15 seconds","description":"Grok Imagine Video 1.5 Lite is the lower-cost tier of xAI's Grok Imagine Video 1.5. It generates video with native audio from a text prompt or from a single starting frame, at 480p, 720p, or 1080p and durations from 1 to 15 seconds. It suits high-volume work such as social clips, ad variations, and fast iteration on prompts and motion.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/xai-grok-imagine-video-1-5-lite.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-imagine-video-1-5-lite.png","pricingOverview":"Pricing starts at $0.02 per second of output at 480p, $0.03 per second of output at 720p, and $0.14 per second of output at 1080p. Starting frame for I2V costs $0.01 per image on top.\n","pricingRates":[{"amount":0.02,"unit":"durationSecond","label":"480p"},{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.14,"unit":"durationSecond","label":"1080p"},{"amount":0.01,"unit":"inputImage"}],"pricingExamples":[{"configuration":"first-frame · 6s","price":"$0.19","amount":0.19,"latencyMs":27355,"exampleId":"dry-break-coupling-product-explainer"},{"configuration":"first-frame · 6s","price":"$0.19","amount":0.19,"latencyMs":30301,"exampleId":"subsea-infrastructure-docuseries-title-teaser"},{"configuration":"text-to-video · 8s","price":"$0.24","amount":0.24,"latencyMs":34078,"exampleId":"algae-bioreactor-laboratory-b-roll"},{"configuration":"text-to-video · 8s","price":"$0.24","amount":0.24,"latencyMs":37150,"exampleId":"bronze-foundry-documentary-title-teaser"}],"guides":[{"slug":"generating-video","title":"Generating video","description":"How to generate video with Grok Imagine Video 1.5 Lite: sizing the request, writing a shot as timed beats, putting a title on screen, and directing the sound.","date":"2026-10-02","url":"https://runware.ai/docs/models/xai-grok-imagine-video-1-5-lite/guides/generating-video"}]},{"model":"recraft-v4-1-flash","air":"recraft:v4.1@flash","name":"Recraft V4.1 Flash","creator":"recraft","status":"live","releasedAt":"2026-09-23T00:00:00Z","weight":4,"headline":"Fast, low-cost design-focused image generation for high-volume work and rapid iteration","description":"Recraft V4.1 Flash is the speed tier of the Recraft V4.1 family, a dedicated fast raster model for text-to-image generation at 1K resolution. It keeps output quality close to Recraft V4.1, with strong results in photography, portraits, atmosphere, complex scenes, mixed media, and logo concepts. It suits high-volume generation, rapid iteration, and design exploration where latency and cost matter most.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-1-flash.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-1-flash.png","pricingRates":[{"amount":0.007,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.007","amount":0.007,"latencyMs":3695,"exampleId":"wildlife-culvert-retrofit-explainer"},{"configuration":"text-to-image","price":"$0.007","amount":0.007,"latencyMs":3851,"exampleId":"deep-sea-robotics-engineer-portrait"},{"configuration":"text-to-image","price":"$0.007","amount":0.007,"latencyMs":3412,"exampleId":"circular-brickmaker-editorial-portrait"},{"configuration":"text-to-image","price":"$0.007","amount":0.007,"latencyMs":3327,"exampleId":"rare-book-conservator-profile-avatar"}]},{"model":"prunaai-p-video-2","air":"prunaai:p-video@2","name":"P-Video-2","creator":"prunaai","status":"live","releasedAt":"2026-09-10T00:00:00Z","weight":4,"headline":"Quality-focused multimodal video generation with native audio, stronger lip sync, and consistent subjects","description":"P-Video-2 is Pruna AI's quality-focused successor to P-Video, creating clips from text, images, or audio-conditioned inputs through one model. It produces sharper close-ups and foreground subjects, stronger identity and input-image consistency, improved lip synchronization for natively generated speech, and clearer on-screen text while generating dialogue, music, and sound effects with the video. It supports optional first-and-last-frame guidance, draft iteration, durations up to 20 seconds, 720p or 1080p output, and 24 or 48 FPS.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-2.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-2.png","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"720p · Standard"},{"amount":0.015,"unit":"durationSecond","label":"720p · Draft"},{"amount":0.05,"unit":"durationSecond","label":"1080p · Standard"},{"amount":0.03,"unit":"durationSecond","label":"1080p · Draft"}],"pricingExamples":[{"configuration":"draft-preview · 720p · 4s","price":"$0.0606","amount":0.0606,"latencyMs":15515,"exampleId":"tomato-grafting-tool-demo-clip"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.201","amount":0.2011,"latencyMs":20233,"exampleId":"mechanical-watch-assembly-b-roll"},{"configuration":"first-frame · 1080p · 5s","price":"$0.252","amount":0.2521,"latencyMs":72176,"exampleId":"maritime-archaeology-logo-ident"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.302","amount":0.3021,"latencyMs":64833,"exampleId":"arctic-seed-vault-presenter"},{"configuration":"audio-driven-image-animation · 1080p","price":"$0.802","amount":0.8021,"latencyMs":106366,"exampleId":"lithium-battery-safety-avatar-clip"},{"configuration":"audio-driven · 1080p","price":"$1.00","amount":1.0027,"latencyMs":103287,"exampleId":"doorway-flood-barrier-explainer-clip"}],"guides":[{"slug":"prompting","title":"Writing prompts for scene, motion and sound","description":"How to write prompts for Pruna P-Video-2, from the clause order the model reads to the framing, camera motion and sound directions that ship inside the same string.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/prompting"},{"slug":"draft-mode","title":"Draft mode and reproducible reruns","description":"How to use settings.draft in Pruna P-Video-2 to test a direction cheaply, what it changes besides quality, and why an approved draft cannot be promoted to a final.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/draft-mode"},{"slug":"resolution-and-duration","title":"Resolution, aspect ratio, frame rate and length","description":"How to size a Pruna P-Video-2 clip: the two mutually exclusive ways to state dimensions, the fourteen accepted pairs, what 48 fps buys, and what happens when you omit duration.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/resolution-and-duration"},{"slug":"native-audio","title":"Native audio, dialogue and lip sync","description":"How to direct the sound P-Video-2 generates with the picture: ambience, effects, music beds, and spoken lines that the mouth actually matches.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/native-audio"},{"slug":"first-and-last-frame","title":"Pinning the first and last frame","description":"How to give Pruna P-Video-2 both ends of a shot, how far apart the two stills can sit before the transition breaks, and the pairing trick that produces a seamless loop.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/first-and-last-frame"},{"slug":"image-to-video","title":"Animating a still with a first frame","description":"How to drive Pruna P-Video-2 from an image with inputs.frameImages, why the still and not your parameters decides the output shape, and how to hold a product across a variant set.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/image-to-video"},{"slug":"audio-driven","title":"Driving a clip from your own audio","description":"How to condition Pruna P-Video-2 on an imported track with inputs.audio, why the audio and not duration sets the length, and when to import a voice instead of generating one.","date":"2026-09-10","url":"https://runware.ai/docs/models/prunaai-p-video-2/guides/audio-driven"}]},{"model":"microsoft-mai-image-2-6-flash","air":"microsoft:mai-image@2.6-flash","name":"MAI-Image-2.6-Flash","creator":"microsoft","status":"coming-soon","releasedAt":"2026-09-04T00:00:00Z","weight":4,"headline":"Fast, production-ready image generation and editing for interactive and high-volume creative workflows","description":"MAI-Image-2.6-Flash is the speed-optimized variant of Microsoft's MAI-Image-2.6 family, retaining its generation and editing capabilities for latency-sensitive production. It supports text-to-image generation, precise multi-reference editing, web grounding, and dynamic aspect ratios, with strengths in targeted object edits, layout adaptation, text updates, artifact cleanup, and visual consistency across iterations. It is designed for rapid prototyping, personalization, automated pipelines, interactive applications, and high-throughput creative production.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/microsoft-mai-image-2-6-flash.jpg","ogImage":"https://assets.runware.ai/og/microsoft-mai-image-2-6-flash.png"},{"model":"prunaai-p-video-edit","air":"prunaai:p-video@edit","name":"P-Video-Edit","creator":"prunaai","status":"live","releasedAt":"2026-09-03T00:00:00Z","weight":4,"headline":"Instruction-based video editing with reference-guided control and preserved motion, structure, and audio","description":"P-Video-Edit is Pruna AI's instruction-based video editing model for changing an existing clip while preserving its motion, scene structure, and audio. Give it a source video and a text instruction to make attribute, object, or environment edits; up to four reference images can guide identity, style, or other supported changes. A draft mode supports faster iteration before rendering the final edit at full quality.","capabilities":["io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-edit.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-edit.png","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"Draft"},{"amount":0.045,"unit":"durationSecond","label":"Standard"}],"pricingExamples":[{"configuration":"reference-to-video","price":"$0.204","amount":0.2038,"latencyMs":103751,"exampleId":"pearlescent-fencing-mask-colorway-reel"},{"configuration":"reference-to-video","price":"$0.204","amount":0.2038,"latencyMs":111728,"exampleId":"seaweed-hatchery-inspection-b-roll"},{"configuration":"video-edit","price":"$0.204","amount":0.2038,"latencyMs":89611,"exampleId":"sage-espresso-machine-colorway-film"},{"configuration":"video-edit","price":"$0.204","amount":0.2038,"latencyMs":93043,"exampleId":"architectural-prosthetic-cover-showcase-reel"}],"guides":[{"slug":"editing-video","title":"Editing video","description":"How to edit a finished clip with Pruna P-Video-Edit: one source video, one written instruction, and the motion, framing and audio the model holds while it changes what you named.","date":"2026-09-03","url":"https://runware.ai/docs/models/prunaai-p-video-edit/guides/editing-video"},{"slug":"draft-mode","title":"Draft mode and reproducible reruns","description":"How to use settings.draft in Pruna P-Video-Edit to preview an edit at a lower rate before committing, and how seed makes the run you approved repeatable.","date":"2026-09-03","url":"https://runware.ai/docs/models/prunaai-p-video-edit/guides/draft-mode"},{"slug":"prompt-upsampling","title":"Prompt upsampling","description":"How settings.promptUpsampling rewrites your instruction before Pruna P-Video-Edit runs it, what it invents when a prompt is thin, and why a precise instruction barely notices it.","date":"2026-09-03","url":"https://runware.ai/docs/models/prunaai-p-video-edit/guides/prompt-upsampling"},{"slug":"reference-images","title":"Editing with reference images","description":"How to steer a Pruna P-Video-Edit edit with up to four reference images, when a still pins down a product or garment better than words, and how references differ from P-Video-Replace.","date":"2026-09-03","url":"https://runware.ai/docs/models/prunaai-p-video-edit/guides/reference-images"},{"slug":"removing-objects","title":"Removing objects from video","description":"How to remove an object or a person from a clip with Pruna P-Video-Edit, naming the target and the surface that continues behind it so the fill holds for every frame.","date":"2026-09-03","url":"https://runware.ai/docs/models/prunaai-p-video-edit/guides/removing-objects"}]},{"model":"bfl-flux-3-upscaler","air":"bfl:flux@video-upscale","name":"FLUX 3 Upscaler","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-08-20T00:00:00Z","weight":4,"headline":"FLUX-powered video upscaling with precise and creative enhancement modes","description":"FLUX 3 Upscaler is a dedicated video enhancement model from Black Forest Labs for increasing source footage to 1080p, 2K, or 4K while preserving motion and the original audio track. It supports a source-faithful precise mode for faces, products, and brand-sensitive footage, plus a more creative enhancement mode that restores and invents finer detail for generated clips, scenery, textures, and crowd shots. The endpoint accepts clips up to 20 seconds long and can optionally use a prompt to steer the type of detail it enhances.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-video-upscale.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-3-upscaler.png","pricingOverview":"Precise: $0.075/Megapixel-second\nCreative: $0.105/Megapixel-second","pricingRates":[{"amount":0.15,"unit":"durationSecond","label":"Precise · 1080p"},{"amount":0.26,"unit":"durationSecond","label":"Precise · 2K"},{"amount":0.59,"unit":"durationSecond","label":"Precise · 4K"},{"amount":0.21,"unit":"durationSecond","label":"Creative · 1080p"},{"amount":0.37,"unit":"durationSecond","label":"Creative · 2K"},{"amount":0.83,"unit":"durationSecond","label":"Creative · 4K"}],"pricingExamples":[{"configuration":"upscale","price":"$2.23","amount":2.2333,"latencyMs":192690,"exampleId":"marquetry-workshop-archive-restoration"},{"configuration":"upscale","price":"$2.23","amount":2.2333,"latencyMs":113288,"exampleId":"subsea-cable-docuseries-titles"},{"configuration":"upscale","price":"$2.23","amount":2.2333,"latencyMs":110844,"exampleId":"bunraku-puppeteer-editorial-portrait"},{"configuration":"upscale","price":"$2.23","amount":2.2333,"latencyMs":187539,"exampleId":"mycelium-leather-lab-b-roll"}],"guides":[{"slug":"upscaling-video","title":"Upscaling video","description":"How to upscale video with FLUX 3 Upscaler: the async request, choosing an upscale factor, the input size and length limits, and keeping motion and the original audio.","date":"2026-08-19","url":"https://runware.ai/docs/models/bfl-flux-3-upscaler/guides/upscaling-video"},{"slug":"prompt-guided-enhancement","title":"Steering the enhancement with a prompt","description":"How to steer FLUX 3 Upscaler with an optional positivePrompt: name the materials and textures you want rebuilt so the enhancement favours the detail that matters.","date":"2026-08-21","url":"https://runware.ai/docs/models/bfl-flux-3-upscaler/guides/prompt-guided-enhancement"},{"slug":"creativity-modes","title":"Precise and creative enhancement","description":"How to use FLUX 3 Upscaler's settings.creativity: precise, source-faithful sharpening for faces and brands, or creative detail restoration for scenery, textures, and crowds.","date":"2026-08-20","url":"https://runware.ai/docs/models/bfl-flux-3-upscaler/guides/creativity-modes"}]},{"model":"recraft-v4-styles","air":"recraft:v4@styles","name":"Recraft V4 Styles","creator":"recraft","status":"live","releasedAt":"2026-08-13T00:00:00Z","weight":4,"headline":"Style-consistent raster generation from reusable reference-based styles without training","description":"Recraft V4 Styles is the standard raster model in Recraft's V4 Styles family for reference-driven style consistency. It lets teams create a reusable visual style from reference images and then generate new assets that stay close not only to the broad aesthetic direction, but also to finer details such as texture, composition, and rendering character. It is designed for workflows that need immediate style reuse without fine-tuning or LoRA setup, making it a strong fit for branded content, editorial systems, product campaigns, and any image pipeline where a repeatable visual language matters more than one-off prompt creativity.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-styles.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-styles.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":25144,"exampleId":"nocturnal-navigation-magazine-cover"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":26377,"exampleId":"cold-storage-field-recording-cover"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":113088,"exampleId":"marine-archaeologist-personal-portrait"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":26178,"exampleId":"artisan-perfumer-editorial-press-portrait"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt Recraft V4 Styles: how references pin the style, what the prompt controls, dimensions, and patterns for consistent output.","date":"2026-08-26","url":"https://runware.ai/docs/models/recraft-v4-styles/guides/prompting"},{"slug":"styles-from-references","title":"Styles from references","description":"How to build a reusable style in Recraft V4 Styles: reference set strategy, styleId reuse across many generations, and same-family portability across the raster and vector tiers.","date":"2026-08-26","url":"https://runware.ai/docs/models/recraft-v4-styles/guides/styles-from-references"},{"slug":"precise-vs-flexible","title":"Precise vs flexible style match","description":"How to pick between Recraft V4 Styles match modes: precise for strict style adherence, flexible for a looser vibe, with side-by-side comparisons.","date":"2026-08-26","url":"https://runware.ai/docs/models/recraft-v4-styles/guides/precise-vs-flexible"}]},{"model":"recraft-v4-styles-pro","air":"recraft:v4@styles-pro","name":"Recraft V4 Styles Pro","creator":"recraft","status":"live","releasedAt":"2026-08-13T00:00:00Z","weight":4,"headline":"Higher-fidelity style-consistent raster generation for premium branded and editorial output","description":"Recraft V4 Styles Pro is the higher-quality raster model in the V4 Styles family. It is built for the same reference-driven style consistency workflow as the standard Styles tier, but is the stronger fit when the output needs more polish, finer detail retention, and more presentation-ready final imagery. Like the rest of the family, it creates reusable styles from reference images without training and aims to preserve specific rendering cues, textures, and compositional behavior across generations rather than only matching the general look of the references.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-styles-pro.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-styles-pro.png","pricingRates":[{"amount":0.105,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.105","amount":0.105,"latencyMs":34190,"exampleId":"cold-storage-retrofit-campaign-visual"},{"configuration":"reference-to-image","price":"$0.105","amount":0.105,"latencyMs":37611,"exampleId":"fermentation-science-quarterly-cover"},{"configuration":"reference-to-image","price":"$0.105","amount":0.105,"latencyMs":36291,"exampleId":"aquaphonic-music-festival-poster"},{"configuration":"reference-to-image","price":"$0.105","amount":0.105,"latencyMs":34896,"exampleId":"pediatric-hospital-picturebook-spread"}]},{"model":"recraft-v4-styles-vector","air":"recraft:v4@styles-vector","name":"Recraft V4 Styles Vector","creator":"recraft","status":"live","releasedAt":"2026-08-13T00:00:00Z","weight":4,"headline":"Style-consistent SVG generation from reusable reference-based design styles","description":"Recraft V4 Styles Vector applies the V4 Styles family to vector output, generating SVG assets that stay aligned to a reusable reference-driven visual style. It is intended for teams that want style consistency not just in raster campaigns, but also in scalable design systems such as icons, logos, graphic marks, packaging elements, and illustration assets. The model is designed to capture the distinctive visual language of reference images without requiring a separate training workflow.","capabilities":["io:text-to-image","io:image-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-styles-vector.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-styles-vector.png","pricingRates":[{"amount":0.055,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.055","amount":0.055,"latencyMs":28675,"exampleId":"kelp-seed-to-sea-process-infographic"},{"configuration":"vectorize","price":"$0.055","amount":0.055,"latencyMs":35850,"exampleId":"arctic-radio-drama-series-poster"},{"configuration":"vectorize","price":"$0.055","amount":0.055,"latencyMs":30846,"exampleId":"artisan-violin-rosin-tin-label"},{"configuration":"vectorize","price":"$0.055","amount":0.055,"latencyMs":26550,"exampleId":"ultrasonic-bat-detector-packshot"}]},{"model":"topazlabs-wonder-3-5","air":"topazlabs:wonder@3.5","name":"Topaz Labs Wonder 3.5","creator":"topazlabs","zdr":true,"status":"live","releasedAt":"2026-08-11T00:00:00Z","weight":4,"headline":"Generative image upscaling with stronger detail, cleaner noise handling, and better precision on structured graphics","description":"Wonder 3.5 is Topaz Labs' updated generative image enhancement and upscaling model for recovering detail and realism from both high-quality and degraded source images. It builds on Wonder 3 with finer detail generation, fewer repetitive patterns, improved handling of visual noise, and better precision on structured content such as tables, charts, and text-heavy graphics. It is well suited to photo restoration, compressed-image cleanup, creative enlargement, and enhancement workflows that need stronger texture recovery without drifting into overly artificial output.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/topazlabs-wonder-3-5.jpg","ogImage":"https://assets.runware.ai/og/topazlabs-wonder-3-5.png","pricingRates":[{"amount":0.0428,"unit":"output","label":"1 MP output"},{"amount":0.0428,"unit":"output","label":"4 MP output"},{"amount":0.0428,"unit":"output","label":"8 MP output"},{"amount":0.0856,"unit":"output","label":"16 MP output"},{"amount":0.1284,"unit":"output","label":"24 MP output"},{"amount":0.1712,"unit":"output","label":"32 MP output"},{"amount":0.214,"unit":"output","label":"40 MP output"},{"amount":0.2996,"unit":"output","label":"50 MP output"},{"amount":0.3424,"unit":"output","label":"64 MP output"},{"amount":0.5564,"unit":"output","label":"100 MP output"}],"pricingExamples":[{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":23518,"exampleId":"planetarium-director-press-headshot"},{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":15453,"exampleId":"anodized-marine-winch-colorway-master"},{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":15274,"exampleId":"terrazzo-medallion-reference-image-cleanup"},{"configuration":"upscale","price":"$0.0428","amount":0.0428,"latencyMs":10723,"exampleId":"kakapo-habitat-report-opener"}],"guides":[{"slug":"upscaling-images","title":"Upscaling images with Wonder 3.5","description":"How to upscale images with Topaz Wonder 3.5: the request shape, choosing an upscale factor, the input and output size limits, and what generative upscaling recovers.","date":"2026-08-12","url":"https://runware.ai/docs/models/topazlabs-wonder-3-5/guides/upscaling-images"},{"slug":"text-and-graphics","title":"Upscaling text and structured graphics","description":"How to sharpen text, charts, tables, and packaging with Topaz Wonder 3.5, recovering legible type and clean structured graphics from soft or compressed sources.","date":"2026-08-15","url":"https://runware.ai/docs/models/topazlabs-wonder-3-5/guides/text-and-graphics"},{"slug":"enhancement-strength","title":"Controlling detail with enhancement strength","description":"How to control how much detail Topaz Wonder 3.5 rebuilds with settings.enhancementStrength: what low, medium, and high change, and when to dial it down for faces or up for texture.","date":"2026-08-13","url":"https://runware.ai/docs/models/topazlabs-wonder-3-5/guides/enhancement-strength"},{"slug":"film-grain","title":"Adding film grain","description":"How to add film grain to a Wonder 3.5 upscale with settings.grain: the silver, gaussian, and grey grain models and the strength, density, and size controls.","date":"2026-08-16","url":"https://runware.ai/docs/models/topazlabs-wonder-3-5/guides/film-grain"},{"slug":"restoring-photos","title":"Restoring old and degraded photos","description":"How to restore old and degraded photos with Topaz Wonder 3.5: faded scans, over-compressed uploads, noisy low-light shots, and tiny legacy files rebuilt as they upscale.","date":"2026-08-14","url":"https://runware.ai/docs/models/topazlabs-wonder-3-5/guides/restoring-photos"}]},{"model":"mistralai-shieldstral-1-0-3b","air":"mistralai:shieldstral@1.0-3b","name":"Shieldstral 1.0 3B","creator":"mistral-ai","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-08-04T00:00:00Z","weight":4,"headline":"Compact multimodal moderation model that judges text and images against your own policy","description":"Shieldstral 1.0 3B is Mistral AI’s moderation model for text and images. The policy travels in the prompt as plain-language questions, so the same model covers products whose rules differ without retraining or a fixed list of harm categories. At 3B parameters it stays cheap enough to sit in front of every request, and its 8K context window holds the policy alongside the content under review. It fits moderation pipelines, review queues, and screening of user uploads before they reach a generation model.","capabilities":["io:text-to-text","io:image-to-text","op:moderation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mistralai-shieldstral-1-0-3b.jpg","ogImage":"https://assets.runware.ai/og/mistralai-shieldstral-1-0-3b.png","contextWindow":8192,"maxOutputTokens":8192,"pricingBasis":"token","pricingRates":[{"amount":9e-8,"unit":"inputToken"},{"amount":9e-8,"unit":"outputToken"},{"amount":4.5e-8,"unit":"cachedInputToken"}]},{"model":"prunaai-p-image-ideogram","air":"prunaai:p-image@ideogram","name":"P-Image-Ideogram","creator":"prunaai","status":"live","releasedAt":"2026-07-30T00:00:00Z","weight":4,"headline":"Text-to-image generation with four quality modes, strong typography, and structured layout control","description":"P-Image-Ideogram is a Pruna AI text-to-image model built in collaboration with Ideogram for fast, high-quality visual generation across product, people, and brand-oriented creative work. It offers four generation modes, from Very Low to High, so teams can choose the right balance of speed, cost, and image quality for each workflow while retaining strong prompt following and unusually capable text rendering. The model supports both natural-language prompting and more structured JSON-style prompting for tighter control over layout, color palettes, bounding boxes, and typography-heavy compositions.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-image-ideogram.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-image-ideogram.png","pricingRates":[{"amount":0.003,"unit":"output","label":"1K - very low thinking"},{"amount":0.006,"unit":"output","label":"2K - very low thinking"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0225","amount":0.0225,"latencyMs":13939,"exampleId":"childrens-water-infrastructure-book-cover"},{"configuration":"text-to-image","price":"$0.0225","amount":0.0225,"latencyMs":12078,"exampleId":"ceramics-studio-air-monitor-lifestyle"},{"configuration":"text-to-image","price":"$0.0225","amount":0.0225,"latencyMs":12378,"exampleId":"sake-brewery-process-icon-set"},{"configuration":"text-to-image","price":"$0.0225","amount":0.0225,"latencyMs":16427,"exampleId":"stage-backdrop-conservation-book-cover"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to prompt P-Image-Ideogram in both modes: natural language for quick exploration and structured JSON for explicit control over layout, color, and typography.","date":"2026-07-30","url":"https://runware.ai/docs/models/prunaai-p-image-ideogram/guides/prompting"},{"slug":"text-and-design","title":"Text-heavy design assets","description":"How to design posters, packaging, ads, and menus with P-Image-Ideogram, and the workflow that gets on-image copy to render clean and exact.","date":"2026-07-30","url":"https://runware.ai/docs/models/prunaai-p-image-ideogram/guides/text-and-design"}]},{"model":"ideogram-object-remover","air":"ideogram:object-remover@0","name":"Ideogram Object Remover","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-07-29T00:00:00Z","weight":4,"headline":"Mask-based object removal with natural scene reconstruction and no prompt required","description":"Ideogram Object Remover removes selected objects from an image using a supplied mask, then reconstructs the hidden area without requiring any text prompt. It is designed to preserve surrounding composition, lighting, textures, shadows, and reflections while cleaning up unwanted people, products, signage, clutter, or other localized elements. The workflow is straightforward: source image plus mask in, edited image out, making it well suited to production retouching, asset cleanup, catalog preparation, and creative post-processing pipelines.","capabilities":["io:image-to-image","op:edit","op:inpaint","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-object-remover.jpg","ogImage":"https://assets.runware.ai/og/ideogram-object-remover.png","pricingRates":[{"amount":0.03,"unit":"output"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.03","amount":0.03,"latencyMs":9302,"exampleId":"cyanotype-wind-farm-cover-art"},{"configuration":"inpaint","price":"$0.03","amount":0.03,"latencyMs":8951,"exampleId":"coffee-packaging-texture-cleanup"},{"configuration":"inpaint","price":"$0.03","amount":0.03,"latencyMs":8346,"exampleId":"museum-gallery-photograph-cleanup"},{"configuration":"inpaint","price":"$0.03","amount":0.03,"latencyMs":10137,"exampleId":"woodblock-oyster-farm-editorial"}],"guides":[{"slug":"removing-objects","title":"Removing objects from images","description":"How to erase objects, people, watermarks, and timestamps from an image with Ideogram Object Remover using a mask, with no prompt required.","date":"2026-07-29","url":"https://runware.ai/docs/models/ideogram-object-remover/guides/removing-objects"}]},{"model":"runware-flux-1-dev-style-lora-training","air":"runware:flux-1-dev@style-lora-training","name":"FLUX.1 [dev] Style LoRA Training","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-24T00:00:00Z","weight":4,"headline":"Style LoRA training workflow for FLUX.1 [dev] with downloadable safetensors output","description":"FLUX.1 [dev] Style LoRA Training is a Runware training workflow for producing style-focused LoRA adapters on top of the FLUX.1 [dev] architecture. It takes a zipped dataset together with a trigger word and training settings, then outputs a LoRA as a safetensors file that can be downloaded or uploaded directly to the platform for immediate use in downstream image generation workflows.","architecture":"flux-1-dev","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/runware-flux-1-dev-style-lora-training.jpg","ogImage":"https://assets.runware.ai/og/runware-flux-1-dev-style-lora-training.png","pricingOverview":"Pricing is $0.00145 per step","pricingRates":[{"amount":0.00145,"unit":"step"}]},{"model":"runware-flux-2-klein-4b-style-lora-training","air":"runware:flux-2-klein-4b@style-lora-training","name":"FLUX.2 [klein] 4B Style LoRA Training","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-24T00:00:00Z","weight":4,"headline":"Style LoRA training workflow for FLUX.2 [klein] 4B with downloadable safetensors output","description":"FLUX.2 [klein] 4B Style LoRA Training is a Runware training workflow for producing style-focused LoRA adapters on top of the FLUX.2 Klein 4B architecture. It takes a zipped dataset together with a trigger word and training settings, then outputs a LoRA as a safetensors file that can be downloaded or uploaded directly to the platform for immediate use in downstream image generation workflows.","architecture":"flux-2-klein-4b","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/runware-flux-2-klein-4b-style-lora-training.jpg","ogImage":"https://assets.runware.ai/og/runware-flux-2-klein-4b-style-lora-training.png","pricingOverview":"Pricing is $0.0035 per step","pricingRates":[{"amount":0.0035,"unit":"step"}]},{"model":"runware-flux-2-klein-9b-style-lora-training","air":"runware:flux-2-klein-9b@style-lora-training","name":"FLUX.2 [klein] 9B Style LoRA Training","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-24T00:00:00Z","weight":4,"headline":"Style LoRA training workflow for FLUX.2 [klein] 9B with downloadable safetensors output","description":"FLUX.2 [klein] 9B Style LoRA Training is a Runware training workflow for producing style-focused LoRA adapters on top of the FLUX.2 Klein 9B architecture. It takes a zipped dataset together with a trigger word and training settings, then outputs a LoRA as a safetensors file that can be downloaded or uploaded directly to the platform for immediate use in downstream image generation workflows.","architecture":"flux-2-klein-9b","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/runware-flux-2-klein-9b-style-lora-training.jpg","ogImage":"https://assets.runware.ai/og/runware-flux-2-klein-9b-style-lora-training.png","pricingOverview":"Pricing is $0.00301 per step","pricingRates":[{"amount":0.00301,"unit":"step"}]},{"model":"runware-qwen-image-style-lora-training","air":"runware:qwen-image@style-lora-training","name":"Qwen-Image Style LoRA Training","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-24T00:00:00Z","weight":4,"headline":"Style LoRA training workflow for Qwen-Image with downloadable safetensors output","description":"Qwen-Image Style LoRA Training is a Runware training workflow for producing style-focused LoRA adapters on top of the Qwen-Image architecture. It uses a zipped training dataset, a trigger word, and configurable training settings to produce a LoRA safetensors file that can either be downloaded or imported automatically into the platform for later image generation use.","architecture":"qwen-image","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/runware-qwen-image-style-lora-training.jpg","ogImage":"https://assets.runware.ai/og/runware-qwen-image-style-lora-training.png","pricingOverview":"Pricing is $0.00105 per step","pricingRates":[{"amount":0.00105,"unit":"step"}]},{"model":"runware-z-image-style-lora-training","air":"runware:z-image@style-lora-training","name":"Z-Image Style LoRA Training","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-24T00:00:00Z","weight":4,"headline":"Style LoRA training workflow for Z-Image with downloadable safetensors output","description":"Z-Image Style LoRA Training is a Runware training workflow for producing style-focused LoRA adapters on top of the Z-Image architecture. It takes a zipped dataset together with a trigger word and training settings, then outputs a LoRA as a safetensors file that can be downloaded or uploaded directly to the platform for immediate use in downstream image generation workflows.","architecture":"z-image","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/runware-z-image-style-lora-training.jpg","ogImage":"https://assets.runware.ai/og/runware-z-image-style-lora-training.png","pricingOverview":"Pricing is $0.00182 per step","pricingRates":[{"amount":0.00182,"unit":"step"}]},{"model":"krea-2-turbo","air":"krea:krea@2-turbo","name":"Krea 2 Turbo","creator":"krea","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-07-14T00:00:00Z","weight":4,"headline":"Fast open-source Krea image model for low-cost text-to-image generation and LoRA-friendly workflows","description":"Krea 2 Turbo is the open-source Krea image endpoint exposed on Runware for fast, low-cost text-to-image generation. It focuses on prompt-driven generation rather than the richer Krea 2 reference workflow, making it a better fit for simple production pipelines, accelerated runtimes, and workflows that benefit from open-model compatibility such as LoRA-based customization.","architecture":"krea-2-turbo-oss","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-2-turbo.jpg","ogImage":"https://assets.runware.ai/og/krea-2-turbo.png","pricingRates":[{"amount":0.015,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":4283,"exampleId":"mine-water-heating-explainer"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":3959,"exampleId":"converted-water-tower-living-room"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":4813,"exampleId":"nitrate-film-preservation-blog-hero"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":4335,"exampleId":"coastal-engineering-book-cover"}]},{"model":"openai-gpt-5-6-luna","air":"openai:gpt@5.6-luna","name":"GPT-5.6 Luna","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-07-09T00:00:00Z","weight":4,"headline":"Cost-efficient GPT-5.6 variant for fast, high-volume text and vision workflows","description":"GPT-5.6 Luna is the most cost-efficient model in OpenAI's GPT-5.6 family, aimed at high-volume and cost-sensitive workloads. It supports a 1,050,000 token context window, up to 128,000 output tokens, image input, structured outputs, function calling, and broad tool use for classification, extraction, summarization, routing, tagging, and lightweight agent tasks.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-6-luna.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-6-luna.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":2e-7,"unit":"inputToken"},{"amount":0.0000012,"unit":"outputToken"},{"amount":2e-8,"unit":"cachedInputToken"}]},{"model":"openai-gpt-5-6-sol","air":"openai:gpt@5.6-sol","name":"GPT-5.6 Sol","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-07-09T00:00:00Z","weight":4,"headline":"Frontier reasoning LLM for complex professional work, long context, and full tool-using workflows","description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 family for complex professional work. It combines the strongest reasoning tier in this family with a 1,050,000 token context window, up to 128,000 output tokens, image input, structured outputs, function calling, and broad tool use for demanding coding, research, analysis, and agent workflows.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-6-sol.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-6-sol.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000004,"unit":"inputToken"},{"amount":0.00002,"unit":"outputToken"},{"amount":4.0000000000000003e-7,"unit":"cachedInputToken"}]},{"model":"openai-gpt-5-6-terra","air":"openai:gpt@5.6-terra","name":"GPT-5.6 Terra","creator":"openai","endpoints":["chat-completions","responses"],"status":"compatible-only","releasedAt":"2026-07-09T00:00:00Z","weight":4,"headline":"Balanced GPT-5.6 model for professional workloads that need strong reasoning with lower cost","description":"GPT-5.6 Terra is the balanced model in OpenAI's GPT-5.6 family, designed for production workloads that need strong reasoning without the cost of the flagship tier. It supports a 1,050,000 token context window, up to 128,000 output tokens, image input, structured outputs, function calling, and broad tool use, making it well suited to coding, analysis, extraction, workflow automation, and general-purpose professional assistants.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-6-terra.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-6-terra.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken"},{"amount":0.000012,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"}]},{"model":"boogu-image-0-1-edit","air":"boogu:image@0.1-edit","name":"Boogu-Image-0.1-Edit","creator":"boogu","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-06-30T00:00:00Z","weight":4,"headline":"Instruction-based image editing with bilingual text rendering","description":"Boogu-Image-0.1-Edit is the instruction editing checkpoint of the 10B Boogu-Image-0.1 family, released under Apache-2.0. It applies a text instruction to a single reference image for object insertion and removal, attribute changes, background replacement and style transfer, and it renders Chinese and English text. The result is returned at the reference image's own size, floored to a multiple of 16 pixels, so the edit keeps the original framing.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/boogu-image-0-1-edit.jpg","ogImage":"https://assets.runware.ai/og/boogu-image-0-1-edit.png","pricingMeasured":[{"configuration":"15 steps · 1024×1024 input","price":0.01198925,"latencyMs":18449},{"configuration":"30 steps · 1024×1024 input","price":0.0239785,"latencyMs":35814},{"configuration":"50 steps · 1024×1024 input","price":0.03937754,"latencyMs":58070}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0323","amount":0.03233798,"latencyMs":59575,"exampleId":"adaptive-raincoat-virtual-try-on"},{"configuration":"reference-to-image","price":"$0.0323","amount":0.03233798,"latencyMs":48596,"exampleId":"conservation-flashlight-editorial-product-scene"},{"configuration":"reference-to-image","price":"$0.0416","amount":0.0415774,"latencyMs":72839,"exampleId":"glassblowing-collective-team-portrait"},{"configuration":"reference-to-image","price":"$0.0416","amount":0.0415774,"latencyMs":77446,"exampleId":"winter-planetarium-campaign-hero"}]},{"model":"boogu-image-0-1-edit-turbo","air":"boogu:image@0.1-edit-turbo","name":"Boogu-Image-0.1-Edit-Turbo","creator":"boogu","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-06-30T00:00:00Z","weight":4,"headline":"Four-step distilled instruction editing for fast turnaround","description":"Boogu-Image-0.1-Edit-Turbo is the four-step distilled edit checkpoint of the 10B Boogu-Image-0.1 family. It applies a text instruction to a single reference image for object insertion and removal, attribute changes, background replacement and style transfer, and it renders Chinese and English text. The result is returned at the reference image's own size.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/boogu-image-0-1-edit-turbo.jpg","ogImage":"https://assets.runware.ai/og/boogu-image-0-1-edit-turbo.png","pricingMeasured":[{"configuration":"4 steps · 1024×1024 input","price":0.0019,"latencyMs":3978}],"pricingExamples":[{"configuration":"reference-to-image · steps 4","price":"$0.0032","amount":0.0032,"latencyMs":6095,"exampleId":"cathedral-conservation-website-hero"},{"configuration":"reference-to-image","price":"$0.0038","amount":0.0038,"latencyMs":7132,"exampleId":"orchard-robot-campaign-photograph"},{"configuration":"reference-to-image","price":"$0.0038","amount":0.0038,"latencyMs":6904,"exampleId":"icelandic-saddle-lifestyle-photograph"},{"configuration":"reference-to-image","price":"$0.0038","amount":0.0038,"latencyMs":6798,"exampleId":"museum-textile-catalog-cleanup"}]},{"model":"bytedance-seedance-2-0-mini","air":"bytedance:seedance@2.0-mini","name":"Seedance 2.0 Mini","creator":"bytedance","status":"live","releasedAt":"2026-06-23T00:00:00Z","weight":4,"headline":"Compact Seedance 2.0 variant for faster, lighter multimodal video generation workflows","description":"Seedance 2.0 Mini is a lighter variant in ByteDance's Seedance 2.0 video model family. It is positioned for teams that want the cinematic prompt understanding and multimodal generation style of Seedance 2.0 in a smaller, more iteration-friendly package. It is suited to fast concepting, quicker turnaround, and broader throughput when the full flagship model is not required.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-2-0-mini.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-2-0-mini.png","pricingOverview":"Pricing is $0.036 per second of video at 480p and $0.081 per second of video at 720p for text/image to video.\nPricing is $0.043 per second of video at 480p and $0.096 per second of video at 720p for video to video.\n","pricingRates":[{"amount":0.036,"unit":"durationSecond","label":"480p · text/image to video"},{"amount":0.081,"unit":"durationSecond","label":"720p · text/image to video"},{"amount":0.043,"unit":"durationSecond","label":"480p · video to video"},{"amount":0.10637,"unit":"durationSecond","label":"720p · video to video"}],"pricingExamples":[{"configuration":"first-last-frame · 480p · 8s","price":"$0.303","amount":0.3028875,"latencyMs":92394,"exampleId":"flood-forecasting-platform-logo-ident"},{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.408","amount":0.408375,"latencyMs":106283,"exampleId":"hospital-pneumatic-tube-logistics-b-roll"},{"configuration":"first-frame · 720p · 6s · audio","price":"$0.489","amount":0.489375,"latencyMs":110386,"exampleId":"sculptural-hearing-jewelry-reel"},{"configuration":"reference-to-video · 720p · 10s · audio","price":"$0.813","amount":0.813375,"latencyMs":115065,"exampleId":"planetary-science-curator-update"}]},{"model":"alibaba-happyhorse-1-1","air":"alibaba:happyhorse@1.1","name":"HappyHorse 1.1","creator":"alibaba","status":"live","releasedAt":"2026-06-22T00:00:00Z","weight":4,"headline":"Multimodal video generation with stronger motion expressiveness, multi-image reference consistency, and improved audio-visual sync","description":"HappyHorse 1.1 is Alibaba's upgraded multimodal video model for text-to-video, image-to-video, and reference-to-video generation. It improves motion continuity, prompt following, character consistency, facial texture quality, cinematic shot logic, and audio-visual synchronization over HappyHorse 1.0, making it better suited to multi-shot storytelling, multi-character scenes, close-up performance, and reference-driven production workflows.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-happyhorse-1-1.jpg","ogImage":"https://assets.runware.ai/og/alibaba-happyhorse-1-1.png","pricingOverview":"Pricing starts at $0.14 per second of video at 720p. Which means $0.42 for the minimum 3 seconds output. Pricing starts at $0.18 per second of video at 1080p. Which means $0.54 for the minimum 3 seconds output.","pricingRates":[{"amount":0.14,"unit":"durationSecond","label":"720p"},{"amount":0.18,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"text-to-video · 1080p · 5s","price":"$0.929","amount":0.9293,"latencyMs":108928,"exampleId":"seaweed-bioplastic-extrusion-b-roll"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.929","amount":0.9293,"latencyMs":112634,"exampleId":"bat-acoustics-network-logo-loop"},{"configuration":"first-frame · 1080p · 8s","price":"$1.46","amount":1.4621,"latencyMs":141416,"exampleId":"volcanology-heat-suit-showcase-reel"},{"configuration":"reference-to-video · 720p · 12s","price":"$1.70","amount":1.6972,"latencyMs":156929,"exampleId":"after-hours-laundromat-music-video-segment"}],"guides":[{"slug":"cinematic-storyboarding","title":"Cinematic shot direction with storyboard prompts","description":"How to write multi-shot storyboard prompts for HappyHorse 1.1 to direct cinematic sequences with shot sizes, camera movement, and subject continuity in a single call.","date":"2026-06-22","url":"https://runware.ai/docs/models/alibaba-happyhorse-1-1/guides/cinematic-storyboarding"},{"slug":"multi-character-casting","title":"Casting multiple characters with reference images","description":"How to use HappyHorse 1.1's reference workflow to cast one or more characters into a generated video and preserve their identity through every cut.","date":"2026-06-22","url":"https://runware.ai/docs/models/alibaba-happyhorse-1-1/guides/multi-character-casting"}]},{"model":"krea-2-raw","air":"krea:krea@2-raw","name":"Krea 2 Raw","creator":"krea","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-06-22T00:00:00Z","weight":4,"headline":"Base 12B Krea 2 checkpoint for finetuning, post-training, and open-weight text-to-image workflows","description":"Krea 2 Raw is the base release checkpoint in the Krea 2 model family, published as an open-weight 12B text-to-image model before the additional post-training and distillation used for Krea 2 Turbo. Krea describes it as a better fit for finetuning, post-training, and custom domain work than for general inference, making it especially relevant for LoRA training and other open-model customization workflows.","architecture":"krea-2-raw-oss","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-2-raw.jpg","ogImage":"https://assets.runware.ai/og/krea-2-raw.png","pricingRates":[{"amount":0.015,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":35167,"exampleId":"water-tower-listening-lounge"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":21910,"exampleId":"night-pollinator-festival-shirt-graphic"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":34976,"exampleId":"luxury-watchmaker-oiler-kit-packshot"},{"configuration":"text-to-image","price":"$0.015","amount":0.015,"latencyMs":51704,"exampleId":"subsea-cable-repair-video-thumbnail"}]},{"model":"prunaai-p-image-try-on","air":"prunaai:p-image@try-on","name":"P-Image-Try-On","creator":"prunaai","status":"live","releasedAt":"2026-06-18T00:00:00Z","weight":4,"headline":"Virtual try-on image model for dressing a person in one or more reference garments while preserving identity and pose","description":"P-Image-Try-On is an image editing model from Pruna AI for virtual try-on workflows. It takes a source image of a person together with one or more garment reference images and generates a new image in which the person is realistically dressed in the provided garments. The model is built to preserve identity, pose, body shape, and the overall structure of the original image, making it useful for fashion previews, e-commerce visualization, styling exploration, and other garment-transfer workflows.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-image-try-on.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-image-try-on.png","pricingOverview":"$0.015 for the first garment, $0.008 for each additional one","pricingRates":[{"amount":0.015,"unit":"output","label":"First garment"},{"amount":0.023,"unit":"output","label":"2 garments"},{"amount":0.031,"unit":"output","label":"3 garments"}],"pricingExamples":[{"configuration":"virtual-try-on","price":"$0.0069","amount":0.0069,"latencyMs":13978,"exampleId":"polar-expedition-uniform-preview"},{"configuration":"pose-guided-try-on","price":"$0.0069","amount":0.0069,"latencyMs":24925,"exampleId":"pipe-organ-technician-editorial-portrait"},{"configuration":"pose-guided-try-on","price":"$0.0069","amount":0.0069,"latencyMs":22883,"exampleId":"celadon-ceramicist-workwear-colorway"},{"configuration":"virtual-try-on","price":"$0.0093","amount":0.0093,"latencyMs":21406,"exampleId":"lighthouse-costume-continuity-cleanup"}],"guides":[{"slug":"virtual-try-on","title":"Dressing a person in a full outfit","description":"How to dress a person in a full outfit with Pruna P-Image-Try-On, passing each garment as its own reference and matching a target pose.","date":"2026-06-18","url":"https://runware.ai/docs/models/prunaai-p-image-try-on/guides/virtual-try-on"}]},{"model":"klingai-video-3-0-turbo","air":"klingai:kling-video@3.0-turbo","name":"Kling VIDEO 3.0 Turbo","creator":"klingai","status":"live","releasedAt":"2026-06-17T00:00:00Z","weight":4,"headline":"Faster multimodal video generation with stronger prompt adherence, multi-shot consistency, and improved lip sync","description":"Kling VIDEO 3.0 Turbo is a speed-optimized multimodal video generation model in the Kling 3.0 family. It is built for high-volume production workflows that need faster turnaround without giving up stable motion, prompt adherence, multi-shot consistency, or audio-visual alignment. It supports text-to-video and image-to-video generation, with particular emphasis on improved lip-sync quality and more efficient large-scale content creation.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-3-0-turbo.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-3-0-turbo.png","pricingOverview":"Pricing is at $0.112 per second of video at 720p with native audio. Starting at $0.336 for 3 seconds","pricingRates":[{"amount":0.112,"unit":"durationSecond","label":"720p"},{"amount":0.14,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"text-to-video · 1080p · 5s","price":"$0.70","amount":0.7,"latencyMs":136056,"exampleId":"luxury-apiary-outerwear-showcase-reel"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.70","amount":0.7,"latencyMs":84700,"exampleId":"forensic-accounting-docuseries-teaser"},{"configuration":"first-frame · 720p · 8s","price":"$0.896","amount":0.896,"latencyMs":127881,"exampleId":"pipe-freeze-clamp-explainer-clip"},{"configuration":"multi-shot · 960×960 · 12s","price":"$1.34","amount":1.344,"latencyMs":106631,"exampleId":"waterworks-cumbia-music-video"}],"guides":[{"slug":"multi-shot-templating","title":"Multi-shot reels with the Kling 3.0 Turbo shot template","description":"How to use Kling 3.0 Turbo's inline shot-list syntax to direct multi-shot video reels in a single API call, with shot-by-shot timing and prompt control.","date":"2026-06-17","url":"https://runware.ai/docs/models/klingai-video-3-0-turbo/guides/multi-shot-templating"}]},{"model":"sourceful-riverflow-2-5-fast","air":"sourceful:riverflow-2.5@fast","name":"Riverflow 2.5 Fast","creator":"sourceful","status":"live","releasedAt":"2026-06-05T00:00:00Z","weight":4,"headline":"Fast production image generation and editing for brand and packaging workflows","description":"Riverflow 2.5 Fast is the speed-focused variant in Sourceful's Riverflow 2.5 family. It supports both text-to-image and image-to-image workflows through the 2.5 generation endpoints, making it suitable for rapid creative iteration, packaging exploration, branded product visuals, and latency-sensitive production pipelines that still need strong visual control.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/sourceful-riverflow-2-5-fast.jpg","ogImage":"https://assets.runware.ai/og/sourceful-riverflow-2-5-fast.png","pricingOverview":"Pricing starts at $0.021 per image generated. Pricing is dependent on thinking effort (low, medium, high)","pricingBasis":"passthrough","pricingRates":[{"amount":0.021,"unit":"output","label":"1K · Low"},{"amount":0.022,"unit":"output","label":"1K · Medium"},{"amount":0.025,"unit":"output","label":"1K · Fast"}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.0226","amount":0.022638,"latencyMs":26926,"exampleId":"underground-waste-network-explainer"},{"configuration":"transparent-background · 768×1024","price":"$0.0237","amount":0.02367,"latencyMs":33831,"exampleId":"bhutan-cultural-tourism-hero-portrait"},{"configuration":"solid-background · 2048×1536","price":"$0.0239","amount":0.023856,"latencyMs":41277,"exampleId":"polar-night-weather-station-visualization"},{"configuration":"reference-to-image · 2K","price":"$0.059","amount":0.058954,"latencyMs":118587,"exampleId":"forensic-botanist-risograph-avatar"},{"configuration":"custom-font-rendering · 1152×768","price":"$0.10","amount":0.100339,"latencyMs":253875,"exampleId":"railway-archive-restoration-poster"}]},{"model":"sourceful-riverflow-2-5-pro","air":"sourceful:riverflow-2.5@pro","name":"Riverflow 2.5 Pro","creator":"sourceful","status":"live","releasedAt":"2026-06-05T00:00:00Z","weight":4,"headline":"Higher-accuracy production image generation and editing for packaging, branding, and commercial design","description":"Riverflow 2.5 Pro is the higher-capability variant in Sourceful's Riverflow 2.5 family. It supports both text-to-image and image-to-image workflows and is positioned for commercial visual work that needs stronger output quality, tighter brand control, and more dependable production results across packaging, product imagery, advertising, and design-heavy creative pipelines.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/sourceful-riverflow-2-5-pro.jpg","ogImage":"https://assets.runware.ai/og/sourceful-riverflow-2-5-pro.png","pricingOverview":"Pricing starts at $0.235 per image generated. Pricing is dependent on thinking effort (low, medium, high, xhigh)","pricingBasis":"passthrough","pricingRates":[{"amount":0.235,"unit":"output","label":"1K · Low"},{"amount":0.236,"unit":"output","label":"1K · Medium"},{"amount":0.245,"unit":"output","label":"1K · High"},{"amount":0.497,"unit":"output","label":"1K · Extra high"}],"pricingExamples":[{"configuration":"text-to-image · 1248×832","price":"$0.179","amount":0.179383,"latencyMs":158332,"exampleId":"historic-cemetery-flood-adaptation-visualization"},{"configuration":"solid-background · 896×1120","price":"$0.203","amount":0.203,"latencyMs":121478,"exampleId":"savory-tomato-cocktail-menu-portrait"},{"configuration":"transparent-background · 1024×1024","price":"$0.253","amount":0.2526,"latencyMs":150239,"exampleId":"luna-moth-pastry-menu-cutout"},{"configuration":"reference-to-image · 2K","price":"$0.72","amount":0.720001,"latencyMs":179955,"exampleId":"tidal-aquaculture-operations-icon-set"}],"guides":[{"slug":"scoring-rubric","title":"The custom scoring rubric","description":"How to use scoringPrompt and scoringRubric on Sourceful Riverflow 2.5 Pro to drive different production workflows from the same brand inputs.","date":"2026-06-05","url":"https://runware.ai/docs/models/sourceful-riverflow-2-5-pro/guides/scoring-rubric"}]},{"model":"prunaai-p-video-replace","air":"prunaai:p-video@replace","name":"P-Video-Replace","creator":"prunaai","status":"live","releasedAt":"2026-06-04T00:00:00Z","weight":4,"headline":"Character replacement for existing video using a reference image while preserving motion, timing, camera, and scene","description":"P-Video-Replace is a video transformation model that swaps the on-camera character in an existing video with the character from a reference image. It is built to preserve the original motion, timing, camera behavior, lighting, and background while changing who appears in the clip, making it useful for UGC ad variations, content localization, avatar or mascot insertion, and other scalable character-replacement workflows.","capabilities":["io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-replace.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-replace.png","pricingOverview":"Pricing starts at $0.03 per second of output at 720p and $0.06 at 1080p","pricingRates":[{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.06,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"video-edit","price":"$0.362","amount":0.3625,"latencyMs":61504,"exampleId":"coral-dive-computer-colorway-video"},{"configuration":"video-edit","price":"$0.362","amount":0.3625,"latencyMs":182001,"exampleId":"cut-paper-curator-exhibition-teaser"},{"configuration":"video-edit","price":"$0.362","amount":0.3625,"latencyMs":108435,"exampleId":"indoor-mushroom-farm-b-roll"},{"configuration":"video-edit","price":"$0.362","amount":0.3625,"latencyMs":104693,"exampleId":"night-rail-technician-recruitment-clip"}],"guides":[{"slug":"replacing-characters-in-video","title":"Replacing the character in a video","description":"How to use Pruna P-Video-Replace to swap the on-camera character in an existing video with one from a reference image while preserving the original motion, timing, camera, lighting, and audio.","date":"2026-06-04","url":"https://runware.ai/docs/models/prunaai-p-video-replace/guides/replacing-characters-in-video"},{"slug":"product-and-wardrobe-variations","title":"Product and wardrobe variations","description":"How to swap a single on-camera object (a product or a garment) in a source video with Pruna P-Video-Replace, without touching the rest of the frame.","date":"2026-06-04","url":"https://runware.ai/docs/models/prunaai-p-video-replace/guides/product-and-wardrobe-variations"},{"slug":"recasting-iconic-film-scenes","title":"Recasting iconic film scenes","description":"How to recreate iconic film scenes with a text-to-video model, then recast the on-camera character with Pruna P-Video-Replace to drop yourself or any reference into the shot.","date":"2026-06-04","url":"https://runware.ai/docs/models/prunaai-p-video-replace/guides/recasting-iconic-film-scenes"}]},{"model":"bytedance-video-enhancement-pro","air":"bytedance:video-enhancement@pro","name":"BytePlus Video Enhancement Pro","creator":"bytedance","status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Higher-grade video enhancement for stronger restoration, clarity, and premium output workflows","description":"BytePlus Video Enhancement Pro is the higher-tier variant in BytePlus AI MediaKit's video enhancement family. It targets more demanding restoration and output-quality workflows, combining video enhancement and upscaling-oriented processing for cases where cleaner detail recovery, stronger visual refinement, and higher presentation quality matter more than basic throughput.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-video-enhancement-pro.jpg","ogImage":"https://assets.runware.ai/og/bytedance-video-enhancement-pro.png","pricingRates":[{"amount":0.034744,"unit":"durationSecond","label":"240p–720p"},{"amount":0.069589,"unit":"durationSecond","label":"1080p"},{"amount":0.13938,"unit":"durationSecond","label":"2K"},{"amount":0.275138,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"upscale","price":"$0.836","amount":0.83628,"latencyMs":543472,"exampleId":"contemporary-dance-archive-restoration"},{"configuration":"upscale","price":"$0.836","amount":0.83628,"latencyMs":382577,"exampleId":"ama-diver-museum-portrait"},{"configuration":"upscale","price":"$0.836","amount":0.83628,"latencyMs":360773,"exampleId":"tattoo-machine-catalog-product-clip"},{"configuration":"upscale","price":"$0.836","amount":0.83628,"latencyMs":381109,"exampleId":"antique-orrery-auction-product-clip"}],"guides":[{"slug":"upscaling-and-enhancing-video","title":"Restoring and upscaling video","description":"How to clean up and upscale soft, compressed, AI-generated, or archival footage to 4K with BytePlus Video Enhancement Pro, and when to reach for it.","date":"2026-07-08","url":"https://runware.ai/docs/models/bytedance-video-enhancement-pro/guides/upscaling-and-enhancing-video"}]},{"model":"bytedance-video-enhancement-standard","air":"bytedance:video-enhancement@standard","name":"BytePlus Video Enhancement Standard","creator":"bytedance","status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Production video enhancement for cleaner upscaled output and routine restoration workflows","description":"BytePlus Video Enhancement Standard is the standard variant in BytePlus AI MediaKit's video enhancement family. It is designed to improve video clarity and presentation quality through resolution enhancement and restoration-oriented processing, making it suitable for routine quality upgrades, compression cleanup, and scalable video post-processing pipelines.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-video-enhancement-standard.jpg","ogImage":"https://assets.runware.ai/og/bytedance-video-enhancement-standard.png","pricingRates":[{"amount":0.003434,"unit":"durationSecond","label":"240p–720p"},{"amount":0.006969,"unit":"durationSecond","label":"1080p"},{"amount":0.013938,"unit":"durationSecond","label":"2K"},{"amount":0.0275138,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"upscale","price":"$0.0418","amount":0.041814,"latencyMs":167836,"exampleId":"aquaculture-biosecurity-presenter-briefing"},{"configuration":"upscale","price":"$0.0418","amount":0.041814,"latencyMs":150158,"exampleId":"planetarium-projector-maintenance-b-roll"},{"configuration":"upscale","price":"$0.0418","amount":0.041814,"latencyMs":117216,"exampleId":"adaptive-beach-mobility-social-reel"},{"configuration":"upscale","price":"$0.0418","amount":0.041814,"latencyMs":117487,"exampleId":"cryogenic-biobank-retrieval-b-roll"}]},{"model":"ideogram-4-0-remix","air":"ideogram:4@remix","name":"Ideogram 4.0 Remix","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Prompt-guided image remixing with controllable source preservation","description":"Ideogram 4.0 Remix uses an existing image as the starting point for a new generation, letting teams rework visuals while controlling how closely the output stays tied to the source. The documented `image_weight` control shifts the balance between prompt-driven transformation and source-image preservation, making it useful for style exploration, brand-safe iteration, creative variation, and design updates that need to retain composition or subject structure.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-4-0-remix.jpg","ogImage":"https://assets.runware.ai/og/ideogram-4-0-remix.png","pricingOverview":"Pricing is $0.03 per image for Turbo, $0.06 per image for Default, and $0.10 per image for Quality","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.1,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"image-to-image","price":"$0.10","amount":0.1,"latencyMs":54862,"exampleId":"neon-archive-campaign-outpaint"},{"configuration":"image-to-image","price":"$0.10","amount":0.1,"latencyMs":35857,"exampleId":"polar-rov-colorway-variant"},{"configuration":"image-to-image","price":"$0.10","amount":0.1,"latencyMs":36508,"exampleId":"diving-helmet-auction-hero"},{"configuration":"image-to-image","price":"$0.10","amount":0.1,"latencyMs":50141,"exampleId":"gasworks-courtyard-evening-visualization"}]},{"model":"ideogram-4-0q","air":"ideogram:4@q","name":"Ideogram 4.0q","creator":"ideogram","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Quantized open-weight design image model for typography, layout control, and self-hosted generation workflows","description":"Ideogram 4.0q is the quantized open-weight variant of Ideogram 4.0, Ideogram's 9.3B design-focused text-to-image model. The public release is available in quantized checkpoint formats such as nf4 and fp8 while retaining the family's strengths in multilingual text rendering, structured JSON prompting, bounding-box layout control, color-palette guidance, and high-resolution design generation. It is a strong fit for teams that want open-model deployment, self-hosted inference, or customization workflows around Ideogram 4 rather than relying only on the hosted API.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-4-0q.jpg","ogImage":"https://assets.runware.ai/og/ideogram-4-0q.png","pricingOverview":"Pricing is $0.03 per image for Turbo, $0.06 per image for Default, and $0.10 per image for Quality","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.1,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.06","amount":0.06,"latencyMs":64860,"exampleId":"livestock-auction-documentary-series-poster"},{"configuration":"text-to-image · 1536×1024","price":"$0.06","amount":0.06,"latencyMs":26414,"exampleId":"bell-foundry-ambient-album-cover"},{"configuration":"text-to-image · 1536×1024","price":"$0.06","amount":0.06,"latencyMs":31916,"exampleId":"pinball-technician-event-shirt-graphic"},{"configuration":"structured-prompt · 1024×1280","price":"$0.10","amount":0.1,"latencyMs":39639,"exampleId":"nocturnal-pollinator-science-art-print"}]},{"model":"ideogram-remove-background","air":"ideogram:remove-background@0","name":"Ideogram Background Remover","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Generative background removal for clean transparent PNG cutouts with strong edge fidelity","description":"Ideogram Background Remover isolates the foreground subject of an image and returns a transparent PNG with clean edges, preserved fine detail, and minimal haloing or fringing. Ideogram positions it as a generative alternative to classic segmentation-based removers, making it especially strong on hard boundaries such as typography, logos, hair, fur, glass, reflections, and intricate product silhouettes. It is well suited to catalog imagery, design assets, compositing, marketplace listings, and high-volume production workflows that need reliable subject extraction without manual cleanup.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/ideogram-remove-background.jpg","ogImage":"https://assets.runware.ai/og/ideogram-remove-background.png","pricingOverview":"Each request cost $0.01","pricingRates":[{"amount":0.01,"unit":"output"}],"pricingExamples":[{"configuration":"remove-background","price":"$0.01","amount":0.01,"latencyMs":8836,"exampleId":"artisan-eyewear-virtual-try-on-asset"},{"configuration":"remove-background","price":"$0.01","amount":0.01,"latencyMs":31750,"exampleId":"risograph-cellist-campaign-cutout"},{"configuration":"remove-background","price":"$0.01","amount":0.01,"latencyMs":11975,"exampleId":"flood-damaged-radio-operator-cutout"},{"configuration":"remove-background","price":"$0.01","amount":0.01,"latencyMs":10326,"exampleId":"lion-dance-head-cutout"}]},{"model":"krea-2-medium-turbo","air":"krea:krea@2-medium-turbo","name":"Krea 2 Medium Turbo","creator":"krea","status":"live","releasedAt":"2026-06-03T00:00:00Z","weight":4,"headline":"Fast Krea 2 variant with creativity control, moodboards, and reference-aware image workflows","description":"Krea 2 Medium Turbo is the fast Krea 2 variant that keeps the richer Krea creative workflow around moodboards, creativity tuning, and reference-driven generation. It is designed for rapid ideation and iteration-heavy image work while still supporting image-to-image generation, style references, and the broader Krea 2 control surface that teams use for guided exploration.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-2-medium-turbo.jpg","ogImage":"https://assets.runware.ai/og/krea-2-medium-turbo.png","pricingRates":[{"amount":0.015,"unit":"output","label":"Text-to-image"},{"amount":0.0175,"unit":"output","label":"Style references"},{"amount":0.02,"unit":"output","label":"Moodboards"}],"pricingExamples":[{"configuration":"reference-to-image · 928×1152","price":"$0.0175","amount":0.0175,"latencyMs":12703,"exampleId":"ev-battery-recycling-blog-hero"},{"configuration":"reference-to-image · 928×1152","price":"$0.0175","amount":0.0175,"latencyMs":12729,"exampleId":"experimental-tidal-album-cover"},{"configuration":"text-to-image · 1376×768","price":"$0.02","amount":0.02,"latencyMs":13702,"exampleId":"overnight-rail-campaign-key-visual"},{"configuration":"text-to-image · 1376×768","price":"$0.02","amount":0.02,"latencyMs":13100,"exampleId":"rare-books-conservator-press-headshot"}]},{"model":"exactly-illustrative-training","air":"exactly:illustrative@training","name":"Exactly Illustrative Training","creator":"exactly","status":"live","releasedAt":"2026-06-01T00:00:00Z","weight":4,"headline":"Custom illustrative model training for consistent branded imagery","description":"Exactly Illustrative Training is a model training workflow for building custom illustrative models from a company's dataset. It fine-tunes the Exactly Illustrative architecture to capture the style of brand assets, then produces a trained model that can be used for consistent text-to-image and image-to-image generation on the platform.","architecture":"exactly-illustrative","capabilities":["op:train"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-training.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-training.png","pricingRates":[{"amount":13,"unit":"output"}],"guides":[{"slug":"training-a-style-model","title":"Training a custom style model","description":"How to train a custom Exactly Illustrative style model on a brand's visual identity, then use it for consistent text-to-image and image-to-image generation in that look.","date":"2026-06-01","url":"https://runware.ai/docs/models/exactly-illustrative-training/guides/training-a-style-model"}]},{"model":"bfl-flux-virtual-try-on","air":"bfl:flux@vto","name":"FLUX Virtual Try-On","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-05-28T00:00:00Z","weight":4,"headline":"Low-latency virtual try-on for transferring garments onto a person image with strong identity and garment fidelity","description":"FLUX Virtual Try-On is a virtual try-on image editing model from Black Forest Labs that generates apparel try-on results from a person image plus one or more garment references. It is tuned to preserve the subject's face and pose while transferring garments with strong logo, print, stitching, and hardware fidelity, making it suitable for catalog-scale styling, product visualization, outfit transfer, and shopper-facing try-on workflows. It supports multi-garment composition, seeded generation, and output sizes up to 2 megapixels.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-virtual-try-on.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-virtual-try-on.png","pricingOverview":"Pricing starts at $0.0375 for the first input MP, then $0.005 per subsequent input MP. In addition it is priced at $0.005 per output MP","pricingRates":[{"amount":0.0375,"unit":"inputMegapixel","label":"First megapixel in"},{"amount":0.005,"unit":"inputMegapixel","label":"Each further megapixel in","after":1},{"amount":0.005,"unit":"outputMegapixel"}],"pricingExamples":[{"configuration":"virtual-try-on","price":"$0.0475","amount":0.0475,"latencyMs":10506,"exampleId":"rooftop-apiary-uniform-composite"},{"configuration":"virtual-try-on","price":"$0.0475","amount":0.0475,"latencyMs":9027,"exampleId":"lava-tube-surveyor-portrait-composite"},{"configuration":"virtual-try-on · steps 4","price":"$0.0475","amount":0.0475,"latencyMs":9391,"exampleId":"winery-guide-banner-expansion"},{"configuration":"virtual-try-on · steps 4","price":"$0.0625","amount":0.0625,"latencyMs":10108,"exampleId":"opera-tailcoat-costume-fitting"}],"guides":[{"slug":"virtual-try-on","title":"Virtual try-on","description":"How to dress a person in any garment from a reference image with FLUX VTO. The call takes one person photo, one garment photo, and a short prompt.","date":"2026-05-28","url":"https://runware.ai/docs/models/bfl-flux-virtual-try-on/guides/virtual-try-on"}]},{"model":"prunaai-p-video-animate","air":"prunaai:p-video@animate","name":"P-Video-Animate","creator":"prunaai","status":"live","releasedAt":"2026-05-28T00:00:00Z","weight":4,"headline":"Reference-image animation driven by the motion, timing, and camera movement of a source video","description":"P-Video-Animate is a motion-transfer video model that animates a single reference image using a source video as the motion driver. It preserves the original acting, timing, camera movement, and scene structure from the driving clip while restyling the output around the supplied image. It is well suited to UGC ad variations, meme remixes, character or avatar recasting, and other high-volume creative workflows that need fast, repeatable image-to-video animation with strong motion fidelity.","capabilities":["io:image-to-video","io:video-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-animate.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-animate.png","pricingOverview":"Pricing starts at $0.03 per second of output video at 720p and $0.06 per second of output video at 1080p","pricingRates":[{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.06,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"reference-to-video","price":"$0.181","amount":0.1813,"latencyMs":34974,"exampleId":"maritime-museum-night-reel"},{"configuration":"reference-to-video","price":"$0.181","amount":0.1813,"latencyMs":45983,"exampleId":"risograph-botanist-music-video-segment"},{"configuration":"reference-to-video","price":"$0.181","amount":0.1813,"latencyMs":33146,"exampleId":"iridescent-apiary-suit-fashion-reel"},{"configuration":"reference-to-video","price":"$0.181","amount":0.1813,"latencyMs":36856,"exampleId":"cleanroom-transfer-port-explainer"}],"guides":[{"slug":"animating-images-from-video","title":"Animating images with a source video","description":"How to use Pruna P-Video-Animate to bring a still reference image to life by inheriting the motion, timing, and camera move from a source video.","date":"2026-05-28","url":"https://runware.ai/docs/models/prunaai-p-video-animate/guides/animating-images-from-video"}]},{"model":"krea-2-large","air":"krea:krea@2-large","name":"Krea 2 Large","creator":"krea","status":"live","releasedAt":"2026-05-27T00:00:00Z","weight":4,"headline":"Larger Krea 2 variant for rawer, more flexible outputs with stronger photorealism and weighted reference control","description":"Krea 2 Large is the higher-capacity variant in the Krea 2 family, with lighter post-training and a more textured, flexible output character. It is the stronger overall model when a workflow benefits from higher ceiling, stronger photorealism, more raw visual character, and better handling of motion blur, film grain, low dynamic range, and other less polished looks. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-2-large.jpg","ogImage":"https://assets.runware.ai/og/krea-2-large.png","pricingOverview":"Pricing for a single image is $0.060. When using an image style reference, it's $0.065.","pricingRates":[{"amount":0.06,"unit":"output","label":"1024x1024"},{"amount":0.065,"unit":"output","label":"1024x1024 · reference image"}],"pricingExamples":[{"configuration":"reference-to-image · 928×1152","price":"$0.065","amount":0.065,"latencyMs":33507,"exampleId":"saffron-fraud-investigation-hero"},{"configuration":"reference-to-image · 928×1152","price":"$0.065","amount":0.065,"latencyMs":34225,"exampleId":"azores-geothermal-tourism-campaign-portrait"},{"configuration":"text-to-image · 1248×832","price":"$0.07","amount":0.07,"latencyMs":35192,"exampleId":"lighthouse-keeper-retirement-supper-invitation"},{"configuration":"text-to-image · 1248×832","price":"$0.07","amount":0.07,"latencyMs":35395,"exampleId":"carousel-horse-auction-hero-image"}]},{"model":"krea-2-medium","air":"krea:krea@2-medium","name":"Krea 2 Medium","creator":"krea","status":"live","releasedAt":"2026-05-27T00:00:00Z","weight":4,"headline":"Faster Krea 2 variant for stable, consistent generation with controllable prompt strength and weighted reference guidance","description":"Krea 2 Medium is the smaller, faster, and more cost-efficient variant in the Krea 2 family. It applies heavier post-training for especially stable and consistent outputs, making it a strong fit for illustration, anime, painting, graphic design, and other expressive visual workflows. It supports text-to-image and image-to-image generation, prompt interpretation strength through the creativity control, and up to 10 weighted reference images with both positive and negative guidance.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/krea-2-medium.jpg","ogImage":"https://assets.runware.ai/og/krea-2-medium.png","pricingOverview":"Pricing for a single image is $0.030. When using an image style reference, it's $0.035.","pricingRates":[{"amount":0.03,"unit":"output","label":"1024x1024"},{"amount":0.035,"unit":"output","label":"1024x1024 · reference image"}],"pricingExamples":[{"configuration":"reference-to-image · 928×1152","price":"$0.035","amount":0.035,"latencyMs":21556,"exampleId":"hydrothermal-vent-wall-print"},{"configuration":"reference-to-image · 928×1152","price":"$0.035","amount":0.035,"latencyMs":29486,"exampleId":"heritage-seed-bank-report-cover"},{"configuration":"text-to-image · 1376×768","price":"$0.04","amount":0.04,"latencyMs":23383,"exampleId":"orchid-conservatory-gala-invitation"},{"configuration":"text-to-image · 1376×768","price":"$0.04","amount":0.04,"latencyMs":22239,"exampleId":"hydrogen-hydrofoil-launch-key-visual"}],"guides":[{"slug":"creativity-and-style-transfer","title":"Creativity control and style transfer","description":"How to steer Krea 2 with the creativity parameter, weighted style reference images, and moodboards. Together these cover faithful renders through bold reinterpretation.","date":"2026-05-27","url":"https://runware.ai/docs/models/krea-2-medium/guides/creativity-and-style-transfer"}]},{"model":"bfl-flux-erase","air":"bfl:flux@erase","name":"FLUX Erase","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-05-21T00:00:00Z","weight":4,"headline":"Dedicated FLUX object removal for clean, context-aware image cleanup","description":"FLUX Erase is a dedicated image editing model from Black Forest Labs for removing unwanted objects, text, or distractions from an image while reconstructing the surrounding content naturally. It is tuned for erase-style workflows where the goal is not to replace an element with something new, but to leave a clean, coherent result that matches nearby structure, lighting, texture, and scene context.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-erase.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-erase.png","pricingOverview":"Generated image: The first generated megapixel is charged $0.03. Each subsequent megapixel is charged $0.004. Reference image(s): $0.004 for each megapixel. For pricing, resolution is always rounded up to the next megapixel, separately for each reference image and for the generated image. 1 megapixel is counted as 1024x1024 pixels. For multiple reference images, each reference image is counted as 1 megapixel. Images exceeding 4 megapixels are resized to 4 megapixels.","pricingRates":[{"amount":0.03,"unit":"outputMegapixel","label":"First megapixel"},{"amount":0.004,"unit":"outputMegapixel","label":"Each further megapixel","after":1},{"amount":0.004,"unit":"inputMegapixel"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.034","amount":0.034,"latencyMs":10094,"exampleId":"blue-hour-observatory-relight-plate"},{"configuration":"inpaint","price":"$0.034","amount":0.034,"latencyMs":9726,"exampleId":"art-deco-backdrop-clean-plate"},{"configuration":"inpaint","price":"$0.034","amount":0.034,"latencyMs":10890,"exampleId":"winter-lavender-farm-calendar-plate"},{"configuration":"inpaint","price":"$0.042","amount":0.042,"latencyMs":13301,"exampleId":"vacant-aircraft-hangar-leasing-photo"}],"guides":[{"slug":"object-removal","title":"Removing objects","description":"How to remove objects from images with FLUX Erase, Black Forest Labs's prompt-less mask-driven removal model. What you paint is what disappears.","date":"2026-05-22","url":"https://runware.ai/docs/models/bfl-flux-erase/guides/object-removal"}]},{"model":"bfl-flux-outpainting","air":"bfl:flux@outpainting","name":"FLUX Outpainting","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2026-05-14T00:00:00Z","weight":4,"headline":"Single-call FLUX outpainting for seamless image expansion","description":"FLUX Outpainting is a dedicated image expansion model from Black Forest Labs that extends an existing image beyond its original borders in a single call. It is tuned to continue scene content, lighting, texture, and composition naturally without requiring a text prompt, making it useful for aspect-ratio changes, social reformats, banner layouts, and other canvas-extension workflows.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-outpainting.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-outpainting.png","pricingOverview":"Pricing is $0.10 for each output megapixel. Output megapixel are defined by the height and width parameter in the API request. Input megapixels are not counted. For pricing, resolution is always rounded up to the next megapixel for the generated image. 1 megapixel is counted as 1024x1024 pixels. Output megapixels are determined by the output width and height. Images exceeding 4 megapixels are resized to 4 megapixels.","pricingRates":[{"amount":0.1,"unit":"outputMegapixel","label":"High · first megapixel"},{"amount":0.1,"unit":"outputMegapixel","label":"High · each further megapixel","after":1},{"amount":0.045,"unit":"outputMegapixel","label":"Fast · first megapixel"},{"amount":0.006,"unit":"outputMegapixel","label":"Fast · each further megapixel","after":1}],"pricingExamples":[{"configuration":"expand","price":"$0.30","amount":0.3,"latencyMs":68439,"exampleId":"equestrian-fly-sheet-lifestyle-hero"},{"configuration":"expand","price":"$0.30","amount":0.3,"latencyMs":96386,"exampleId":"rooftop-apiary-crew-portrait"},{"configuration":"expand","price":"$0.30","amount":0.3,"latencyMs":68952,"exampleId":"mokuhanga-kelp-restoration-report-cover"},{"configuration":"expand","price":"$0.40","amount":0.4,"latencyMs":146504,"exampleId":"field-microscope-colorway-catalog-hero"}],"guides":[{"slug":"extending-images","title":"Extending images","description":"How to extend an image past its original frame with FLUX Outpainting. No prompt, no mask, just more canvas around what is already there.","date":"2026-05-22","url":"https://runware.ai/docs/models/bfl-flux-outpainting/guides/extending-images"}]},{"model":"recraft-v4-1","air":"recraft:v4.1@0","name":"Recraft V4.1","creator":"recraft","status":"live","releasedAt":"2026-05-14T00:00:00Z","weight":4,"headline":"Sharper design-focused image generation for everyday creative work and iteration","description":"Recraft V4.1 is the standard raster model in the Recraft V4.1 family for professional image generation and editing. It improves the V4 line with cleaner photorealism, sharper object understanding, smoother gradients and 3D rendering, cleaner icons and vectors by default, and better results from shorter prompts, while staying faster and more cost-efficient than the Pro variant.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-1.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-1.png","pricingRates":[{"amount":0.035,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":7839,"exampleId":"animated-adventure-streaming-poster"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":15755,"exampleId":"period-conspiracy-series-poster"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":8621,"exampleId":"geothermal-field-crew-sticker-pack"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":11139,"exampleId":"artisan-glassworks-game-key-art"}],"guides":[{"slug":"prompting","title":"Prompting for photorealism, illustration, and design","description":"How to write effective prompts for Recraft V4.1, from short interpretive prompts to structured multi-layer descriptions for precise creative control.","date":"2026-05-19","url":"https://runware.ai/docs/models/recraft-v4-1/guides/prompting"},{"slug":"color-palettes","title":"Controlling output colors with palettes and backgrounds","description":"How to use the settings.colors and settings.backgroundColor parameters to lock generated images to a specific color palette or background color.","date":"2026-05-19","url":"https://runware.ai/docs/models/recraft-v4-1/guides/color-palettes"}]},{"model":"recraft-v4-1-pro","air":"recraft:v4.1-pro@0","name":"Recraft V4.1 Pro","creator":"recraft","status":"live","releasedAt":"2026-05-14T00:00:00Z","weight":4,"headline":"Higher-resolution design-focused image generation for premium and print-ready output","description":"Recraft V4.1 Pro is the higher-resolution raster model in the Recraft V4.1 family for premium creative production. It shares the same improved design taste and capabilities as V4.1, including cleaner photorealism, stronger object understanding, smoother gradients and 3D rendering, cleaner icon and logo output, and better short-prompt behavior, but is tuned for larger and more polished final assets.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-1-pro.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-1-pro.png","pricingRates":[{"amount":0.21,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":18331,"exampleId":"volcanologist-personal-brand-portrait"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":17678,"exampleId":"marine-coating-campaign-key-visual"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":13831,"exampleId":"residential-geothermal-social-media-ad"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":12096,"exampleId":"bank-vault-listening-lounge-interior"}]},{"model":"recraft-v4-1-utility","air":"recraft:v4.1-utility@0","name":"Recraft V4.1 Utility","creator":"recraft","status":"live","releasedAt":"2026-05-14T00:00:00Z","weight":4,"headline":"Controlled image generation for product shots, icons, and predictable commercial visuals","description":"Recraft V4.1 Utility is the controlled-output raster model in the Recraft V4.1 family. It is designed for cases where predictability matters more than art direction, with flatter lighting, front-facing composition, and simpler scenes that make it a strong fit for product shots, icons, e-commerce imagery, and other clean commercial visuals.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-1-utility.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-1-utility.png","pricingRates":[{"amount":0.035,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":11894,"exampleId":"crop-insurance-social-campaign-visual"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":12173,"exampleId":"oceanographic-vessel-launch-invitation"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":10493,"exampleId":"urban-beekeeper-personal-brand-portrait"},{"configuration":"text-to-image","price":"$0.035","amount":0.035,"latencyMs":16786,"exampleId":"pipe-organ-craft-journal-cover"}]},{"model":"recraft-v4-1-utility-pro","air":"recraft:v4.1-utility-pro@0","name":"Recraft V4.1 Utility Pro","creator":"recraft","status":"live","releasedAt":"2026-05-14T00:00:00Z","weight":4,"headline":"Higher-resolution controlled image generation for polished product and e-commerce assets","description":"Recraft V4.1 Utility Pro is the higher-resolution controlled-output raster model in the Recraft V4.1 family. It is built for predictable, clean commercial imagery with flatter lighting, front-facing compositions, and simpler scenes, making it well suited to polished product shots, icons, packaging visuals, and e-commerce assets that need consistency over creative variation.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-1-utility-pro.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-1-utility-pro.png","pricingRates":[{"amount":0.21,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":17052,"exampleId":"subsea-cable-technician-persona-portrait"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":16170,"exampleId":"space-toilet-science-book-spread"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":25070,"exampleId":"living-shoreline-storm-buffer-infographic"},{"configuration":"text-to-image","price":"$0.21","amount":0.21,"latencyMs":17314,"exampleId":"copper-beekeeping-smoker-product-packshot"}]},{"model":"xai-grok-4-3","air":"xai:grok@4.3","name":"Grok 4.3","creator":"xai","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-05-05T00:00:00Z","weight":4,"headline":"Flagship multimodal LLM for agentic tool calling, long context, and low-hallucination reasoning","description":"Grok 4.3 is xAI's flagship language model for agentic reasoning, strong instruction following, and minimal hallucinations. It supports text and image input, a 1 million token context window, configurable reasoning effort including non-reasoning mode, function calling, and structured outputs for production assistants, coding workflows, and long-context analysis.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/xai-grok-4-3.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-4-3.png","contextWindow":1000000,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":0.00000125,"unit":"inputToken"},{"amount":0.0000025,"unit":"outputToken"},{"amount":2e-7,"unit":"cachedInputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000206","amount":0.0002059,"latencyMs":4469,"exampleId":"newsroom-embargo-timestamp-lookup"},{"configuration":"tool-calling","price":"$0.000236","amount":0.0002359,"latencyMs":4184,"exampleId":"pharmaceutical-recall-timestamp-record"},{"configuration":"text-generation","price":"$0.00265","amount":0.00264935,"latencyMs":14578,"exampleId":"saas-pricing-migration-decision-memo"},{"configuration":"text-generation","price":"$0.00313","amount":0.00312685,"latencyMs":12013,"exampleId":"clinical-trial-consent-rewrite"}]},{"model":"luma-uni-1","air":"luma:uni@1","name":"UNI-1","creator":"luma","status":"live","releasedAt":"2026-05-05T00:00:00Z","weight":4,"headline":"Balanced unified image model for creation, precision editing, and multi-reference control","description":"UNI-1 is a unified image model in Luma's UNI-1 family for both image creation and precision image editing. It combines text prompting, source-image modification, multi-reference guidance, seed-based reproducibility, and reasoning-informed visual generation in one system, with strong control over composition, identity, style, and visual plausibility.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/luma-uni-1.jpg","ogImage":"https://assets.runware.ai/og/luma-uni-1.png","pricingOverview":"Pricing at 2048 pixels for Text to image is $0.0404, for Image edit (image to image) is $0.0434 and $0.003 more for each additional reference image : $0.0464 for 2 ref image, $0.0494 for 3 ref image, etc.","pricingRates":[{"amount":0.0404,"unit":"output","label":"2K"},{"amount":0.003,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1440p","price":"$0.0404","amount":0.0404,"latencyMs":125761,"exampleId":"shinkansen-ekiben-menu-hero"},{"configuration":"text-to-image · 1440p","price":"$0.0404","amount":0.0404,"latencyMs":118817,"exampleId":"big-bud-tractor-landscape-composite"},{"configuration":"image-to-image · 2464×1696","price":"$0.0464","amount":0.0464,"latencyMs":95316,"exampleId":"avalanche-field-recording-album-cover"},{"configuration":"reference-to-image · 1504×2784","price":"$0.0464","amount":0.0464,"latencyMs":83699,"exampleId":"azores-geothermal-hiking-campaign-illustration"}]},{"model":"luma-uni-1-max","air":"luma:uni@1-max","name":"UNI-1 Max","creator":"luma","status":"live","releasedAt":"2026-05-05T00:00:00Z","weight":4,"headline":"Higher-quality unified image model for premium generation and precision editing","description":"UNI-1 Max is the quality-first variant in Luma's UNI-1 image family for both image creation and precision image editing. It uses the same API shape and capability set as UNI-1, but is tuned for higher-quality output when detail, polish, and final-image quality matter more than using the default variant.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/luma-uni-1-max.jpg","ogImage":"https://assets.runware.ai/og/luma-uni-1-max.png","pricingOverview":"Pricing at 2048 pixels for Text to image is $0.1, for Image edit (image to image) is $0.103 and $0.003 more for each additional reference image : $0.106 for 2 ref image, $0.109 for 3 ref image, etc.","pricingRates":[{"amount":0.1,"unit":"output","label":"2K"},{"amount":0.003,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 2784×1504","price":"$0.10","amount":0.1,"latencyMs":222348,"exampleId":"adaptive-commuter-rainwear-lookbook"},{"configuration":"text-to-image · 2784×1504","price":"$0.10","amount":0.1,"latencyMs":165023,"exampleId":"big-boy-locomotive-cutaway-plate"},{"configuration":"reference-to-image · 1696×2464","price":"$0.106","amount":0.106,"latencyMs":146072,"exampleId":"tidal-energy-manga-cover-expansion"},{"configuration":"image-to-image · 2048×2048","price":"$0.109","amount":0.109,"latencyMs":188130,"exampleId":"vieques-bioluminescent-bay-campaign"}]},{"model":"heygen-avatar-v","air":"heygen:avatar@5","name":"HeyGen Avatar V","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-05-04T00:00:00Z","weight":4,"headline":"High-fidelity avatar video generation with stronger identity consistency and lip sync","description":"HeyGen Avatar V is an avatar video generation model for talking digital twins and other eligible registered avatar looks. It improves identity preservation, lip sync accuracy, facial expressiveness, and motion coherence across angle changes, scene changes, and long-form videos, making it well suited to presenter, training, and localization workflows where avatar stability matters.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/heygen-avatar-v.jpg","ogImage":"https://assets.runware.ai/og/heygen-avatar-v.png","pricingOverview":"Each generation will cost $0.1/s","pricingRates":[{"amount":0.12,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"background-replacement · 1080p","price":"$1.14","amount":1.1446,"latencyMs":202726,"exampleId":"rare-book-handling-stock-clip"},{"configuration":"captioned-stock-avatar · 720p","price":"$2.64","amount":2.6448,"latencyMs":540258,"exampleId":"funeral-home-mystery-teaser"},{"configuration":"scripted-stock-avatar · 1080p","price":"$2.71","amount":2.712,"latencyMs":186091,"exampleId":"heritage-documentary-series-teaser"},{"configuration":"audio-driven-stock-avatar · 1080p","price":"$3.35","amount":3.3512,"latencyMs":157910,"exampleId":"draft-line-hygiene-meter-demo"}],"guides":[{"slug":"script-vs-audio","title":"Driving the avatar: text to speech or your own audio","description":"How to choose between Avatar V's two input modes: generate the voice from a script, or drive the avatar with your own recorded audio.","date":"2026-05-20","url":"https://runware.ai/docs/models/heygen-avatar-v/guides/script-vs-audio"},{"slug":"backgrounds-and-framing","title":"Backgrounds, framing, and aspect ratios","description":"How to compose what surrounds the avatar in Avatar V output: the background, fit mode, aspect ratio for the target platform, and burned-in captions.","date":"2026-05-20","url":"https://runware.ai/docs/models/heygen-avatar-v/guides/backgrounds-and-framing"}]},{"model":"prunaai-p-video-avatar","air":"prunaai:p-video@avatar","name":"P-Video-Avatar","creator":"prunaai","status":"live","releasedAt":"2026-04-30T00:00:00Z","weight":4,"headline":"High-fidelity talking avatar video generation from a portrait with script or audio control","description":"P-Video-Avatar is a portrait-driven avatar video model that turns a single image into a speaking video using either an uploaded audio track or a generated voice from script. It is built for production avatar workflows with strong lip sync, selectable voices and languages, optional speaking-style control, seeded generation, and 720p or 1080p output for scalable talking-head video creation.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video-avatar.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video-avatar.png","pricingOverview":"0.025$ per second of output video at 720p and 0.045$ per second of output video at 1080p","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"720p"},{"amount":0.045,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"scripted-photo-avatar · 720p","price":"$0.219","amount":0.219,"latencyMs":26242,"exampleId":"indigo-dyehouse-music-video-bridge"},{"configuration":"scripted-photo-avatar · 720p","price":"$0.423","amount":0.423,"latencyMs":247988,"exampleId":"airport-biosecurity-training-presenter"},{"configuration":"audio-driven-photo-avatar · 1080p","price":"$0.488","amount":0.4878,"latencyMs":64941,"exampleId":"prestige-fragrance-mystery-teaser"},{"configuration":"audio-driven-photo-avatar · 1080p","price":"$0.50","amount":0.5004,"latencyMs":65880,"exampleId":"seed-vault-thriller-teaser"}]},{"model":"rundiffusion-juggernaut-z","air":"rundiffusion:200@100","name":"Juggernaut Z","creator":"rundiffusion","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-29T00:00:00Z","weight":4,"headline":"Polished image model with stronger cinematic lighting, cleaner focus, and richer portrait detail","description":"Juggernaut Z is a fine-tuned image model built on Z Image Base and tuned for a more presentation-ready visual style. Its strengths include more dramatic lighting, stronger camera focus, more detailed skin textures, cleaner anatomy, improved scene composition, and more coherent rendering across portraits, fashion, architecture, and cinematic concept work.","architecture":"z-image","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/rundiffusion-juggernaut-z.jpg","ogImage":"https://assets.runware.ai/og/rundiffusion-juggernaut-z.png","pricingOverview":"Price depends on acceleration parameter. None: $ 0.01495, Low: $0.0117, Medium: $0.01248","pricingMeasured":[{"configuration":"1024 x 1024","price":0.01495}],"pricingExamples":[{"configuration":"image-to-image · 1280×768 · steps 40","price":"$0.00338","amount":0.00338,"latencyMs":5349,"exampleId":"planetarium-anniversary-digital-invitation"},{"configuration":"text-to-image · 1024×1024 · steps 35","price":"$0.00416","amount":0.00416,"latencyMs":5324,"exampleId":"apiary-cooperative-team-portrait-composite"},{"configuration":"inpaint · 768×1024","price":"$0.00416","amount":0.00416,"latencyMs":7123,"exampleId":"urban-mushroom-farming-blog-hero"},{"configuration":"expand · 720p","price":"$0.00494","amount":0.00494,"latencyMs":8111,"exampleId":"converted-lifeboat-station-listing-interior"}]},{"model":"alibaba-happyhorse-1-0","air":"alibaba:happyhorse@1.0","name":"HappyHorse-1.0","creator":"alibaba","status":"live","releasedAt":"2026-04-27T00:00:00Z","weight":4,"headline":"Text-to-video and image-to-video model with 1080p output and short-form clip control","description":"HappyHorse-1.0 is a video generation model for text-to-video and image-to-video workflows. It supports output at 720p or 1080p, clip durations from 3 to 15 seconds, seeded generation, watermark control, and first-frame image conditioning for image-to-video generation.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-happyhorse-1-0.jpg","ogImage":"https://assets.runware.ai/og/alibaba-happyhorse-1-0.png","pricingOverview":"text-to-video, image-to-video and ref-to-video is $0.14 / second (720p) and $0.24 / second (1080p). For video edit price is $0.14 / second (720p) and $0.24 / second (1080p) and both input and output video seconds will be charged.\n","pricingRates":[{"amount":0.24,"unit":"durationSecond","label":"1080p"},{"amount":0.14,"unit":"durationSecond","label":"720p"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s","price":"$0.723","amount":0.7228,"latencyMs":96671,"exampleId":"mycelium-packaging-drop-test-reel"},{"configuration":"reference-to-video · 832×1088 · 6s","price":"$0.857","amount":0.8572,"latencyMs":110838,"exampleId":"manuscript-pigment-scanner-product-demo"},{"configuration":"text-to-video · 1080p · 5s","price":"$1.24","amount":1.239,"latencyMs":154272,"exampleId":"coral-restoration-cleanup-visualization"},{"configuration":"video-edit · 5s · audio origin","price":"$1.45","amount":1.4544,"latencyMs":140405,"exampleId":"aviation-volcanologist-moving-headshot"}]},{"model":"skywork-skyreels-v4","air":"skywork:skyreels@v4","name":"SkyReels V4","creator":"skywork","status":"live","releasedAt":"2026-04-24T00:00:00Z","weight":4,"headline":"Unified multimodal video model for generation, inpainting, and editing with synchronized audio","description":"SkyReels V4 is a unified multimodal video foundation model for joint video-audio generation, inpainting, and editing. It accepts text, images, video clips, masks, and audio references, and supports cinematic outputs up to 1080p, 32 FPS, and 15 seconds with synchronized audio, making it suitable for prompt-driven generation as well as guided editing workflows.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/skywork-skyreels-v4.jpg","ogImage":"https://assets.runware.ai/og/skywork-skyreels-v4.png","pricingRates":[{"amount":0.11,"unit":"durationSecond","label":"Text/image to video · 480p"},{"amount":0.14,"unit":"durationSecond","label":"Text/image to video · 720p"},{"amount":0.35,"unit":"durationSecond","label":"Text/image to video · 1080p"},{"amount":0.18,"unit":"durationSecond","label":"Video to video · 480p"},{"amount":0.25,"unit":"durationSecond","label":"Video to video · 720p"},{"amount":0.625,"unit":"durationSecond","label":"Video to video · 1080p"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s · audio","price":"$0.704","amount":0.7044,"latencyMs":145885,"exampleId":"colorized-lighthouse-heritage-clip"},{"configuration":"multiple-keyframes · 480p · 9s","price":"$0.994","amount":0.9935,"latencyMs":243216,"exampleId":"volcanologist-keynote-motion-headshot"},{"configuration":"audio-driven · 720p · 8s","price":"$1.12","amount":1.1244,"latencyMs":537609,"exampleId":"vertical-seed-vault-technician-clip"},{"configuration":"video-extend · 720p · 5s","price":"$1.26","amount":1.2579,"latencyMs":567834,"exampleId":"antique-auction-video-cleanup"},{"configuration":"first-last-frame · 1080p · 6s","price":"$2.11","amount":2.111,"latencyMs":226331,"exampleId":"vintage-neon-restoration-social-reel"},{"configuration":"text-to-video · 1080p · 6s · audio","price":"$2.11","amount":2.111,"latencyMs":224099,"exampleId":"subsea-cable-documentary-teaser"},{"configuration":"reference-to-video · 1080p · 8s","price":"$3.77","amount":3.7697,"latencyMs":534926,"exampleId":"court-stenographer-thriller-title-sequence"}]},{"model":"klingai-video-3-0-4k","air":"klingai:kling-video@3-4k","name":"Kling VIDEO 3.0 4K","creator":"klingai","status":"live","releasedAt":"2026-04-23T00:00:00Z","weight":4,"headline":"4K multimodal video generation with native audio, multi-shot narratives, and stronger visual detail","description":"Kling VIDEO 3.0 4K is the 4K variant of Kling VIDEO 3.0 for text-to-video and image-to-video generation. It extends the 3.0 series from 720p Standard and 1080p Pro into 4K output while keeping the same multimodal strengths: native audio generation, multi-shot sequencing, element consistency, prompt-driven scene control, and stable temporal coherence across longer clips.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-3-0-4k.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-3-0-4k.png","pricingRates":[{"amount":0.42,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-video · 2160p · 5s","price":"$2.10","amount":2.1,"latencyMs":118496,"exampleId":"oyster-farm-workwear-showcase-reel"},{"configuration":"first-frame · 7s","price":"$2.94","amount":2.94,"latencyMs":197738,"exampleId":"urban-bat-survey-logo-loop"},{"configuration":"first-last-frame · 8s","price":"$3.36","amount":3.36,"latencyMs":183555,"exampleId":"water-service-line-explainer"},{"configuration":"multi-shot · 2880×2880 · 9s","price":"$3.78","amount":3.78,"latencyMs":179275,"exampleId":"meteorite-conservation-curator-social-briefing"},{"configuration":"elements · 10s","price":"$4.20","amount":4.2,"latencyMs":326482,"exampleId":"sleeper-train-cabin-reel"}]},{"model":"pixverse-modify","air":"pixverse:modify@0","name":"PixVerse Modify","creator":"pixverse","status":"live","releasedAt":"2026-04-15T00:00:00Z","weight":4,"headline":"Mask-aware video editing for swaps, removals, restyling, and prompt-driven scene changes","description":"PixVerse Modify is a video-to-video editing model for changing existing footage with text instructions, optional reference images, and masks. It supports subject swapping, object addition and removal, free-form scene edits such as weather or lighting changes, in-video text replacement, and full-video style transfer while preserving the source clip structure.","capabilities":["io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-modify.jpg","ogImage":"https://assets.runware.ai/og/pixverse-modify.png","pricingOverview":"Pricing starts at $0.04/s at 360p, $0.05/s at 540p, and $0.06/s at 720p.","pricingRates":[{"amount":0.04,"unit":"durationSecond","label":"360p"},{"amount":0.05,"unit":"durationSecond","label":"540p"},{"amount":0.06,"unit":"durationSecond","label":"720p"}],"pricingExamples":[{"configuration":"inpaint · 360p","price":"$0.28","amount":0.28,"latencyMs":81966,"exampleId":"vaccine-cooler-field-placement-film"},{"configuration":"reference-to-video · 540p","price":"$0.35","amount":0.35,"latencyMs":146800,"exampleId":"sculptural-floor-lamp-placement-film"},{"configuration":"video-edit · 720p","price":"$0.42","amount":0.42,"latencyMs":173819,"exampleId":"badlands-fossil-exhibit-promo"},{"configuration":"video-edit · 720p","price":"$0.42","amount":0.42,"latencyMs":218654,"exampleId":"embroidered-garment-workshop-music-video"}]},{"model":"minimax-music-2-6","air":"minimax:music@2.6","name":"MiniMax Music 2.6","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-04-10T00:00:00Z","weight":4,"headline":"Promptable full-song generation with vocals, lyrics, BPM and key control","description":"MiniMax Music 2.6 is MiniMax’s latest music generation model for full vocal songs and instrumentals from text prompts. It supports natural-language prompts or detailed production-style instructions, follows specified BPM and key with high reliability, and exposes fine-grained song structure control through section tags. The same Music API also supports instrumental generation, lyrics-assisted workflows, and synchronous or streaming delivery.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-music-2-6.jpg","ogImage":"https://assets.runware.ai/og/minimax-music-2-6.png","pricingOverview":" $0.15 per song generation or per music cover","pricingRates":[{"amount":0.15,"unit":"output"}],"pricingExamples":[{"configuration":"lyrics-to-music","price":"$0.15","amount":0.15,"latencyMs":141145,"exampleId":"womens-rugby-championship-anthem"},{"configuration":"optimized-lyrics-to-music","price":"$0.15","amount":0.15,"latencyMs":102113,"exampleId":"manuscript-handling-microlearning-song"},{"configuration":"text-to-audio","price":"$0.15","amount":0.15,"latencyMs":151506,"exampleId":"community-seed-library-folk-anthem"},{"configuration":"instrumental-music","price":"$0.15","amount":0.15,"latencyMs":140747,"exampleId":"urban-infrastructure-podcast-theme"}]},{"model":"minimax-music-cover","air":"minimax:music@cover","name":"MiniMax Music Cover","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-04-10T00:00:00Z","weight":4,"headline":"Audio-to-audio song transformation that preserves melody while changing style","description":"MiniMax Music Cover is MiniMax’s song-to-song transformation model for reimagining an existing track in a new style. It preserves the original vocal melody while changing voice timbre, instrumentation, genre, and arrangement through a text prompt. It supports one-step generation from reference audio or a two-step workflow with preprocessing and optional lyric editing.","capabilities":["io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-music-cover.jpg","ogImage":"https://assets.runware.ai/og/minimax-music-cover.png","pricingRates":[{"amount":0.15,"unit":"output"}],"pricingExamples":[{"configuration":"audio-driven","price":"$0.15","amount":0.15,"latencyMs":181750,"exampleId":"electric-ferry-boarding-jingle"},{"configuration":"audio-driven","price":"$0.15","amount":0.15,"latencyMs":143138,"exampleId":"plant-patent-podcast-opening-theme"},{"configuration":"audio-driven","price":"$0.15","amount":0.15,"latencyMs":112959,"exampleId":"wood-frog-audiobook-learning-interlude"},{"configuration":"audio-driven","price":"$0.15","amount":0.15,"latencyMs":100465,"exampleId":"abyssal-observatory-ambient-bed"}]},{"model":"bytedance-seedance-2-0-fast","air":"bytedance:seedance@2.0-fast","name":"Seedance 2.0 Fast","creator":"bytedance","status":"live","releasedAt":"2026-04-03T00:00:00Z","weight":4,"headline":"Faster multimodal video generation tuned for latency and iteration speed","description":"Seedance 2.0 Fast is a speed-optimized variant of ByteDance's unified multimodal audio-video generation model. It accepts text, image, audio, and video inputs in combination, like Seedance 2.0, but targets shorter wall-clock times and higher throughput for iterative workflows. It produces multi-shot videos with dual-channel synchronized audio including dialogue, ambient sound, and effects, with physics-aware motion and editing controls, while prioritizing responsiveness over the last increment of visual refinement so teams can preview and ship ideas faster.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-2-0-fast.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-2-0-fast.png","pricingOverview":"Text/Image to Video, 480p: $0.06/sec\nText/Image to Video, 720p: $0.13/sec\nVideo-to-Video, 480p: from $0.10/sec\nVideo-to-Video, 720p: from $0.21/sec","pricingRates":[{"amount":0.06,"unit":"durationSecond","label":"480p"},{"amount":0.13,"unit":"durationSecond","label":"720p"},{"amount":0.16695,"unit":"durationSecond","label":"720p · video to video"}],"pricingExamples":[{"configuration":"first-last-frame · 480p · 8s · audio","price":"$0.485","amount":0.48462,"latencyMs":104843,"exampleId":"wheelchair-fencing-arena-background-swap"},{"configuration":"first-frame · 720p · 5s","price":"$0.653","amount":0.6534,"latencyMs":136593,"exampleId":"alpaca-ranch-background-replacement"},{"configuration":"text-to-video · 1470×630 · 6s · audio","price":"$0.787","amount":0.786822,"latencyMs":94347,"exampleId":"subsea-drone-background-replacement"},{"configuration":"reference-to-video · 720p · 10s · audio","price":"$1.29","amount":1.28835,"latencyMs":251797,"exampleId":"counterfeit-currency-docuseries-titles"}]},{"model":"alibaba-wan2-7","air":"alibaba:wan@2.7","name":"Wan2.7","creator":"alibaba","status":"live","releasedAt":"2026-04-03T00:00:00Z","weight":4,"headline":"Multimodal video generation with reference consistency, video editing, and native audio","description":"Wan2.7 is Alibaba's next-generation multimodal video model supporting text-to-video, image-to-video, reference-to-video, and video editing. It features multi-shot storytelling, subject-consistent multi-character generation, first-and-last-frame interpolation, video continuation, style transfer, instruction-based editing, and audio-conditioned generation with auto-dubbing. Output at 720p or 1080p, 30 FPS in multiple aspect ratios.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-7.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-7.png","pricingOverview":"Both input (if any) and output video seconds will be charged at $0.10 per second for 720p and $0.15 per second for 1080p","pricingRates":[{"amount":0.1,"unit":"durationSecond","label":"720p"},{"amount":0.15,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 720p · 7s · audio","price":"$0.708","amount":0.7082,"latencyMs":66369,"exampleId":"night-shift-lyric-video-segment"},{"configuration":"first-frame · 1080p · 5s · audio","price":"$0.756","amount":0.7557,"latencyMs":51401,"exampleId":"urban-beekeeping-suit-showcase-reel"},{"configuration":"audio-driven · 720p · 8s","price":"$0.803","amount":0.8034,"latencyMs":187284,"exampleId":"ice-core-archive-safety-presenter"},{"configuration":"reference-to-video · 1088×832 · 10s · audio","price":"$1.00","amount":1.0031,"latencyMs":399542,"exampleId":"natural-indigo-workshop-b-roll"},{"configuration":"text-to-video · 1080p · 15s · audio","price":"$2.25","amount":2.2534,"latencyMs":418197,"exampleId":"municipal-noir-limited-series-teaser"}]},{"model":"ace-step-v1-5-xl-base","air":"runware:ace-step@v1.5-xl-base","name":"ACE-Step v1.5 XL Base","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-02T00:00:00Z","weight":4,"headline":"4B music generation model with higher audio quality and full editing task support","description":"ACE-Step v1.5 XL Base is the 4B DiT variant of ACE-Step 1.5 for high-quality music generation and editing. It supports text-to-music, cover generation, repaint, extract, lego, and complete workflows, uses 50 inference steps with CFG, and is designed for longer-form audio generation up to 10 minutes with broad multilingual prompt support.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ace-step-v1-5-xl-base.jpg","ogImage":"https://assets.runware.ai/og/ace-step-v1-5-xl-base.png","pricingMeasured":[{"configuration":"duration 60 · steps 20","price":0.00195,"latencyMs":4667},{"configuration":"duration 30 · steps 50","price":0.00285,"latencyMs":4735},{"configuration":"duration 60 · steps 50","price":0.0039,"latencyMs":6689},{"configuration":"duration 60 · steps 100","price":0.00675,"latencyMs":9363},{"configuration":"duration 180 · steps 50","price":0.0096,"latencyMs":15335},{"configuration":"duration 300 · steps 50","price":0.0153,"latencyMs":32114}],"pricingExamples":[{"configuration":"text-to-audio · 60s","price":"$0.0039","amount":0.0039,"latencyMs":15867,"exampleId":"forensic-mycology-podcast-opening-theme"},{"configuration":"audio-extend","price":"$0.0039","amount":0.0039,"latencyMs":18448,"exampleId":"rare-book-fraud-podcast-intro"},{"configuration":"audio-driven","price":"$0.0057","amount":0.0057,"latencyMs":17286,"exampleId":"seed-vault-alarm-training-narration"},{"configuration":"audio-repaint","price":"$0.00765","amount":0.00765,"latencyMs":22135,"exampleId":"aquatic-veterinary-podcast-case-stinger"}]},{"model":"ace-step-v1-5-xl-sft","air":"runware:ace-step@v1.5-xl-sft","name":"ACE-Step v1.5 XL SFT","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-02T00:00:00Z","weight":4,"headline":"Highest-quality 4B music generation model with CFG-controlled prompt adherence","description":"ACE-Step v1.5 XL SFT is the supervised fine-tuned 4B DiT variant in the ACE-Step 1.5 XL line. It is positioned as the highest-quality XL option, combining 50-step CFG inference with stronger prompt adherence and refined audio quality for text-to-music, cover, and repaint workflows when final output quality matters more than speed or broader editing task coverage.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ace-step-v1-5-xl-sft.jpg","ogImage":"https://assets.runware.ai/og/ace-step-v1-5-xl-sft.png","pricingMeasured":[{"configuration":"duration 60 · steps 20","price":0.00195,"latencyMs":5523},{"configuration":"duration 30 · steps 50","price":0.00285,"latencyMs":4967},{"configuration":"duration 60 · steps 50","price":0.0039,"latencyMs":7831},{"configuration":"duration 60 · steps 100","price":0.00675,"latencyMs":9367}],"pricingExamples":[{"configuration":"text-to-audio · 60s","price":"$0.0039","amount":0.0039,"latencyMs":13205,"exampleId":"retro-tech-podcast-theme-song"},{"configuration":"audio-repaint","price":"$0.0096","amount":0.0096,"latencyMs":24229,"exampleId":"maritime-claims-podcast-segment-bumper"},{"configuration":"audio-extend","price":"$0.0115","amount":0.01155,"latencyMs":30820,"exampleId":"neutrino-observatory-ambient-extension"},{"configuration":"audio-driven","price":"$0.0144","amount":0.0144,"latencyMs":32299,"exampleId":"flooded-archive-salvage-audio-lesson"}]},{"model":"ace-step-v1-5-xl-turbo","air":"runware:ace-step@v1.5-xl-turbo","name":"ACE-Step v1.5 XL Turbo","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-02T00:00:00Z","weight":4,"headline":"Fast 4B music generation model with 8-step inference for higher-quality rapid iteration","description":"ACE-Step v1.5 XL Turbo is the accelerated 4B DiT variant of ACE-Step 1.5. It is optimized for faster music generation with 8-step distilled inference while retaining the higher-capacity XL architecture. It supports text-to-music, cover generation, and repaint workflows, making it suitable for rapid iteration when the 2B turbo model is not enough in audio quality.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ace-step-v1-5-xl-turbo.jpg","ogImage":"https://assets.runware.ai/og/ace-step-v1-5-xl-turbo.png","pricingMeasured":[{"configuration":"duration 60 · steps 8","price":0.0009,"latencyMs":3608},{"configuration":"duration 300 · steps 8","price":0.00285,"latencyMs":10892}],"pricingExamples":[{"configuration":"text-to-audio · 60s","price":"$0.0009","amount":0.0009,"latencyMs":12051,"exampleId":"forensic-pollen-e-learning-song"},{"configuration":"audio-extend","price":"$0.00195","amount":0.00195,"latencyMs":20152,"exampleId":"wastewater-energy-narration-spot"},{"configuration":"audio-repaint","price":"$0.0039","amount":0.0039,"latencyMs":21815,"exampleId":"rare-books-podcast-intro-segment"},{"configuration":"audio-driven","price":"$0.0048","amount":0.0048,"latencyMs":21291,"exampleId":"heritage-crop-seed-vault-voiceover"}]},{"model":"google-gemma-4-31b","air":"google:gemma@4-31b","name":"Gemma 4 31B","creator":"google","hosted":true,"precision":"bf16","zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2026-04-02T00:00:00Z","weight":4,"headline":"Open 31B multimodal reasoning model for coding, long context, and agentic workflows","description":"Gemma 4 31B is Google's flagship dense open-weights model in the Gemma 4 family. It combines strong reasoning, coding performance, native function calling, multimodal understanding across text, image, and video, and a 256K context window in a 31B-parameter open model designed for local and cloud deployment.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemma-4-31b.jpg","ogImage":"https://assets.runware.ai/og/google-gemma-4-31b.png","contextWindow":262144,"maxOutputTokens":262144,"pricingBasis":"token","pricingRates":[{"amount":1.02e-7,"unit":"inputToken"},{"amount":2.97e-7,"unit":"outputToken"},{"amount":1.2e-8,"unit":"cachedInputToken"}]},{"model":"alibaba-wan2-7-image","air":"alibaba:wan@2.7-image","name":"Wan2.7 Image","creator":"alibaba","status":"live","releasedAt":"2026-04-01T00:00:00Z","weight":4,"headline":"Unified image generation and editing with avatar customization, color control, and multilingual text rendering","description":"Wan2.7 Image is a unified image generation and editing model from Alibaba that combines generation and interactive editing in a shared latent space. It features virtual avatar face customization with fine bone structure and eye shape control, a color palette system for extracting and applying consistent color schemes, precise marquee selection editing for pixel-level element manipulation, multilingual text rendering supporting up to 3000 tokens in 12 languages, and compositional generation of up to 12 images in a single output.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-7-image.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-7-image.png","pricingRates":[{"amount":0.03,"unit":"output"}],"pricingExamples":[{"configuration":"sequential-images · 1080p","price":"$0.03","amount":0.03,"latencyMs":13865,"exampleId":"coastal-weather-archive-restoration"},{"configuration":"reference-to-image · 1024×1536","price":"$0.03","amount":0.03,"latencyMs":17501,"exampleId":"community-radio-pledge-drive-post"},{"configuration":"text-to-image · 1536×1024","price":"$0.03","amount":0.03,"latencyMs":5043,"exampleId":"textile-water-recycling-campaign-visual"},{"configuration":"local-editing · 1280×1280","price":"$0.03","amount":0.03,"latencyMs":13283,"exampleId":"volcanologist-risograph-profile-avatar"}]},{"model":"alibaba-wan2-7-image-pro","air":"alibaba:wan@2.7-image-pro","name":"Wan2.7 Image Pro","creator":"alibaba","status":"live","releasedAt":"2026-04-01T00:00:00Z","weight":4,"headline":"Premium image generation with enhanced composition stability and precise prompt comprehension","description":"Wan2.7 Image Pro is the premium variant of Wan2.7 Image offering more stable composition and more precise prompt comprehension. It shares all capabilities of the standard model including avatar customization, color palette control, marquee editing, multilingual text rendering across 12 languages, and multi-image composition, with improved consistency and fidelity for professional workflows.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-7-image-pro.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-7-image-pro.png","pricingRates":[{"amount":0.075,"unit":"output"}],"pricingExamples":[{"configuration":"local-editing · 1024×1024","price":"$0.075","amount":0.075,"latencyMs":15504,"exampleId":"experimental-radio-festival-poster"},{"configuration":"reference-to-image · 1280×1024","price":"$0.075","amount":0.075,"latencyMs":16211,"exampleId":"oyster-hatchery-meter-lifestyle-photograph"},{"configuration":"sequential-images · 1344×768","price":"$0.075","amount":0.075,"latencyMs":17061,"exampleId":"saltworks-visitor-lighting-study-set"},{"configuration":"text-to-image · 1536×1024","price":"$0.075","amount":0.075,"latencyMs":12210,"exampleId":"rare-book-conservator-press-portrait"}]},{"model":"exactly-illustrative-bold-chromatics","air":"exactly:illustrative@bold-chromatics","name":"Exactly Bold Chromatics","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":4,"headline":"Vibrant, high-contrast illustrative style with bold color palettes","description":"Bold Chromatics is an illustrative style from Exactly AI that produces vibrant, high-contrast artwork with striking color palettes. It generates expressive illustrations with bold chromatic choices, ideal for eye-catching visuals, brand illustrations, and creative projects that demand strong visual impact.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-bold-chromatics.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-bold-chromatics.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"pixverse-v6","air":"pixverse:1@8","name":"PixVerse V6","creator":"pixverse","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":4,"headline":"Multi-shot cinematic video generation with native audio, 20+ camera controls, and character consistency","description":"PixVerse V6 is a video generation model focused on multi-shot storytelling with native synchronized audio. It provides over 20 cinematic camera controls including focal length, aperture, depth of field, lens distortion, and vignetting. It features improved character consistency across shots using multi-image references, supports 1080p output at up to 15 seconds, and includes multilingual text rendering in frames.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/pixverse-v6.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v6.png","pricingRates":[{"amount":0.025,"unit":"durationSecond","label":"360p without audio"},{"amount":0.035,"unit":"durationSecond","label":"360p with audio"},{"amount":0.035,"unit":"durationSecond","label":"540p without audio"},{"amount":0.045,"unit":"durationSecond","label":"540p with audio"},{"amount":0.045,"unit":"durationSecond","label":"720p without audio"},{"amount":0.06,"unit":"durationSecond","label":"720p with audio"},{"amount":0.09,"unit":"durationSecond","label":"1080p without audio"},{"amount":0.115,"unit":"durationSecond","label":"1080p with audio"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 5s","price":"$0.225","amount":0.225,"latencyMs":38041,"exampleId":"harbor-pilot-motion-headshot"},{"configuration":"video-edit · 720p","price":"$0.225","amount":0.225,"latencyMs":77773,"exampleId":"cabaret-emcee-motion-portrait-retouch"},{"configuration":"first-last-frame · 540p · 8s","price":"$0.28","amount":0.28,"latencyMs":58678,"exampleId":"luthiery-exhibition-engraving-teaser"},{"configuration":"first-frame · 720p · 6s · audio","price":"$0.36","amount":0.36,"latencyMs":49793,"exampleId":"museum-entomology-set-extension"},{"configuration":"multi-shot · 720p · 10s","price":"$0.45","amount":0.45,"latencyMs":89106,"exampleId":"period-kitchen-background-replacement"}],"guides":[{"slug":"multi-shot-storytelling","title":"Multi-shot storytelling with anchor frames","description":"How to use PixVerse V6 to generate multi-shot video reels from one or two anchor frames in a single call, using the beat-structured prompt pattern.","date":"2026-06-15","url":"https://runware.ai/docs/models/pixverse-v6/guides/multi-shot-storytelling"}]},{"model":"bfl-flux-2-klein-9b-kv","air":"runware:400@6","name":"FLUX.2 [klein] 9B KV","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-03-20T00:00:00Z","weight":4,"headline":"KV-cache accelerated image generation and editing for real-time multi-reference workflows","description":"FLUX.2 [klein] 9B KV is a KV-cache optimized variant of the Klein 9B model that caches reference image key-value pairs after the first denoising step, skipping redundant computation on subsequent steps. This delivers up to 2.5x faster inference for multi-reference editing tasks while retaining all capabilities of the standard Klein 9B, including sub-second text-to-image and advanced editing in 4 steps.","architecture":"flux-2-klein-9b","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-klein-9b-kv.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-klein-9b-kv.png","pricingOverview":"From $0.00078/ image","pricingMeasured":[{"configuration":"1024x1024 · 4 steps","price":0.00078}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.00169","amount":0.00169,"latencyMs":1860,"exampleId":"converted-bank-loft-interior"},{"configuration":"text-to-image · 1024×1024","price":"$0.00169","amount":0.00169,"latencyMs":2064,"exampleId":"flower-market-thriller-series-poster"},{"configuration":"reference-to-image · 1280×768","price":"$0.00247","amount":0.00247,"latencyMs":3997,"exampleId":"childrens-mudskipper-science-spread"},{"configuration":"reference-to-image · 1280×768","price":"$0.00247","amount":0.00247,"latencyMs":4432,"exampleId":"battery-recycling-scientist-editorial-portrait"}]},{"model":"minimax-m2-7","air":"minimax:m2.7@0","name":"MiniMax M2.7","creator":"minimax","endpoints":["chat-completions"],"status":"deprecated","replacedBy":"minimax-m3","deactivatesAt":"2026-11-18T00:00:00Z","releasedAt":"2026-03-18T00:00:00Z","weight":4,"headline":"Long-context agentic coding and office productivity model for fast, reliable tool use","description":"MiniMax M2.7 is a long-context LLM designed for agentic workflows across software engineering, search and tool use, and high-value office productivity tasks. It’s built for multi-step execution, with strong instruction following and dependable task decomposition, making it a solid default for production assistants that write code, call tools, and handle complex document workflows.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-m2-7.jpg","ogImage":"https://assets.runware.ai/og/minimax-m2-7.png","contextWindow":204800,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":3e-7,"unit":"inputToken"},{"amount":0.0000012,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000151","amount":0.000151,"latencyMs":1989,"exampleId":"production-service-status-tool-call"},{"configuration":"tool-calling","price":"$0.000242","amount":0.000242,"latencyMs":2852,"exampleId":"digital-pathology-status-triage"},{"configuration":"text-generation","price":"$0.00389","amount":0.003887,"latencyMs":33134,"exampleId":"historic-theater-grant-application-narrative"},{"configuration":"text-generation","price":"$0.00473","amount":0.00473,"latencyMs":58260,"exampleId":"food-recall-decision-memorandum"}]},{"model":"google-gemini-3-1-flash-lite","air":"google:gemini@3.1-flash-lite","name":"Gemini 3.1 Flash Lite","creator":"google","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-03-03T00:00:00Z","weight":4,"headline":"Advanced multimodal text and reasoning model","description":"Gemini 3.1 Flash Lite is Google’s flagship multimodal language model that processes text alongside images, audio, video, code, and documents. It offers high-performance reasoning, complex instruction following, and deep contextual understanding for a wide range of tasks across language, analysis, and problem solving","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-1-flash-lite.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-1-flash-lite.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":2.5e-7,"unit":"inputToken","label":"text, image, video"},{"amount":5e-7,"unit":"inputToken","label":"audio"},{"amount":0.0000015,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000053","amount":0.000053,"latencyMs":1192,"exampleId":"museum-exhibition-readiness-check"},{"configuration":"text-generation","price":"$0.000652","amount":0.000652,"latencyMs":2323,"exampleId":"cold-chain-incident-briefing"},{"configuration":"audio-driven","price":"$0.00177","amount":0.001768,"latencyMs":6726,"exampleId":"wholesale-buyer-voicemail-action-brief"},{"configuration":"structured-output","price":"$0.00181","amount":0.001807,"latencyMs":3383,"exampleId":"retail-lease-abstract-json-record"},{"configuration":"reference-to-text","price":"$0.00183","amount":0.001833,"latencyMs":11242,"exampleId":"hotel-suite-conformance-report"}]},{"model":"bytedance-seedream-5-0-lite","air":"bytedance:seedream@5.0-lite","name":"Seedream 5.0 Lite","creator":"bytedance","status":"live","releasedAt":"2026-02-24T00:00:00Z","weight":4,"headline":"Responsive text-to-image generation with real-time search and precise prompt adherence","description":"Seedream 5.0 Lite is an advanced image generation model from ByteDance that produces high-quality still images from text prompts while providing flexibility for editing workflows. It is designed to combine expressive creativity with precise control over layout, composition, styles, and details, interpreting nuanced instructions faithfully. Users can incorporate a single reference image to guide generation or editing. Integrated search and reasoning features let the model visualize real-time trends and domain information in the output.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-5-0-lite.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-5-0-lite.png","pricingOverview":"Pricing is $0.035 for both 2K & 3K outputs.","pricingRates":[{"amount":0.035,"unit":"output"}],"pricingExamples":[{"configuration":"sequential-images · 1600×2848","price":"$0.035","amount":0.035,"latencyMs":256432,"exampleId":"four-stage-compost-inoculant-packaging"},{"configuration":"reference-to-image · 2304×1728","price":"$0.035","amount":0.035,"latencyMs":75464,"exampleId":"estuary-sound-festival-poster"},{"configuration":"text-to-image · 2048×2048","price":"$0.035","amount":0.035,"latencyMs":59603,"exampleId":"cryogenic-glove-laboratory-campaign-still"},{"configuration":"text-to-image · 2048×2048","price":"$0.035","amount":0.035,"latencyMs":111457,"exampleId":"oyster-farm-radio-lifestyle-photograph"}]},{"model":"google-gemini-3-1-pro","air":"google:gemini@3.1-pro","name":"Gemini 3.1 Pro","creator":"google","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-02-19T00:00:00Z","weight":4,"headline":"Advanced multimodal text and reasoning model","description":"Gemini 3.1 Pro is Google’s flagship multimodal language model that processes text alongside images, audio, video, code, and documents. It offers high-performance reasoning, complex instruction following, and deep contextual understanding for a wide range of tasks across language, analysis, and problem solving.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-1-pro.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-1-pro.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":0.000002,"unit":"inputToken","label":"<200k"},{"amount":0.000004,"unit":"inputToken","label":">200k"},{"amount":0.000012,"unit":"outputToken","label":"<200k"},{"amount":0.000018,"unit":"outputToken","label":">200k"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.00259","amount":0.002588,"latencyMs":5009,"exampleId":"global-newsroom-embargo-bulletin"},{"configuration":"reference-to-text","price":"$0.0544","amount":0.054366,"latencyMs":53267,"exampleId":"hotel-suite-design-conformance-report"},{"configuration":"audio-driven","price":"$0.0615","amount":0.061538,"latencyMs":44338,"exampleId":"commercial-lease-negotiation-memo"},{"configuration":"text-generation","price":"$0.108","amount":0.108242,"latencyMs":91530,"exampleId":"b2b-saas-pricing-migration-playbook"}]},{"model":"recraft-vectorize","air":"recraft:1@1","name":"Recraft Vectorize","creator":"recraft","status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":4,"headline":"Raster to SVG Vector Conversion for Logos and Icons","description":"Recraft Vectorize converts PNG or JPG artwork into clean SVG vectors. It keeps edges sharp with minimal artifacts. Ideal for logos, icons, UI assets, and flat illustrations. Integrate in pipelines that need scalable graphics from legacy raster files.","capabilities":["io:image-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-vectorize.jpg","ogImage":"https://assets.runware.ai/og/recraft-vectorize.png","pricingRates":[{"amount":0.01,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.01","amount":0.01,"latencyMs":9155,"exampleId":"maritime-museum-diver-screenprint-portrait"},{"configuration":"vectorize","price":"$0.01","amount":0.01,"latencyMs":10110,"exampleId":"winter-lights-glasshouse-map-illustration"},{"configuration":"vectorize","price":"$0.01","amount":0.01,"latencyMs":6698,"exampleId":"rescue-helmet-colorway-vector"},{"configuration":"vectorize","price":"$0.01","amount":0.01,"latencyMs":10277,"exampleId":"updated-antarctic-crew-portrait-vector"}]},{"model":"alibaba-qwen-image-2-0-pro","air":"alibaba:qwen-image@2.0-pro","name":"Qwen-Image-2.0-Pro","creator":"alibaba","status":"live","releasedAt":"2026-02-10T00:00:00Z","weight":4,"headline":"Enhanced image generation and precision editing for professional workflows","description":"Qwen-Image-2.0-Pro builds on Qwen-Image-2.0 with optimized visual fidelity, improved layout and typography handling, and advanced editing control for professional creative and enterprise applications. It delivers richer detail, more accurate text and iconography rendering, and refined editing semantics across a wide range of visual styles, making it suitable for advertising, branding, design systems, and high-impact visual content.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-2-0-pro.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-2-0-pro.png","pricingRates":[{"amount":0.075,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.075","amount":0.075,"latencyMs":17242,"exampleId":"marine-radio-oyster-farm-lifestyle"},{"configuration":"reference-to-image · 1536×1024","price":"$0.075","amount":0.075,"latencyMs":41463,"exampleId":"antarctic-station-crew-photo-edit"},{"configuration":"reference-to-image · 1536×1024","price":"$0.075","amount":0.075,"latencyMs":45436,"exampleId":"solar-filter-kit-packaging"},{"configuration":"text-to-image · 1024×1024","price":"$0.075","amount":0.075,"latencyMs":17512,"exampleId":"ilulissat-winter-tourism-campaign-poster"}]},{"model":"xai-grok-imagine-image","air":"xai:grok-imagine@image","name":"Grok Imagine Image","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-01-29T00:00:00Z","weight":4,"headline":"AI image generation from text and images","description":"Grok Imagine Image is a multimodal generative image model that creates high-quality still images from text prompts or image inputs. It supports flexible visual synthesis across a range of styles, enabling developers to generate creative imagery directly from structured prompts or to expand on existing visuals with coherent, detailed outputs.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/xai-grok-imagine-image.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-imagine-image.png","pricingOverview":"Each image gen costs $0.02 at 1024x1024. (+$0.002 for input images for editing.)","pricingRates":[{"amount":0.02,"unit":"output"},{"amount":0.002,"unit":"inputImage"}],"pricingExamples":[{"configuration":"text-to-image · 1408×768","price":"$0.02","amount":0.02,"latencyMs":9470,"exampleId":"arctic-airlift-strategy-key-art"},{"configuration":"text-to-image · 1408×768","price":"$0.02","amount":0.02,"latencyMs":8384,"exampleId":"oyster-restoration-panoramic-masthead"},{"configuration":"reference-to-image · 1K","price":"$0.022","amount":0.022,"latencyMs":18276,"exampleId":"heritage-hotel-facade-cleanup"},{"configuration":"reference-to-image · 1K","price":"$0.022","amount":0.022,"latencyMs":20174,"exampleId":"period-telephone-thriller-poster"}]},{"model":"xai-grok-imagine-video","air":"xai:grok-imagine@video","name":"Grok Imagine Video","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-01-29T00:00:00Z","weight":4,"headline":"AI video generation with synchronized audio from text and images","description":"Grok Imagine Video is a multimodal generative video model that produces short video clips with native audio from text descriptions or static images. It supports text-to-video and image-to-video generation with synchronized sound effects and dialogue, enabling developers to animate scenes with motion, camera dynamics, and audio in a single API workflow.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/xai-grok-imagine-video.jpg","ogImage":"https://assets.runware.ai/og/xai-grok-imagine-video.png","pricingOverview":"A 6-second T2V video starts at $0.30; each extra second is $0.05 (+$0.002 if using an image input or +$0.01/s of video input).","pricingRates":[{"amount":0.3,"unit":"output","label":"480p · text to video · up to 6s"},{"amount":0.42,"unit":"output","label":"720p · text to video · up to 6s"},{"amount":0.302,"unit":"output","label":"480p · image to video · up to 6s"},{"amount":0.422,"unit":"output","label":"720p · image to video · up to 6s"},{"amount":0.36,"unit":"output","label":"480p · video to video · up to 6s"},{"amount":0.48,"unit":"output","label":"720p · video to video · up to 6s"},{"amount":0.05,"unit":"durationSecond","after":6,"label":"480p · each second beyond the first six"},{"amount":0.07,"unit":"durationSecond","after":6,"label":"720p · each second beyond the first six"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s","price":"$0.352","amount":0.352,"latencyMs":71430,"exampleId":"automated-car-wash-music-segment"},{"configuration":"text-to-video · 720p · 6s","price":"$0.42","amount":0.42,"latencyMs":63880,"exampleId":"period-railway-platform-clean-plate"},{"configuration":"video-edit · 6s","price":"$0.482","amount":0.4825,"latencyMs":80456,"exampleId":"museum-pest-training-avatar-clip"},{"configuration":"reference-to-video · 720p · 8s","price":"$0.564","amount":0.564,"latencyMs":53370,"exampleId":"industrial-bakery-music-video-segment"},{"configuration":"video-extend · 8s","price":"$0.621","amount":0.6205,"latencyMs":125899,"exampleId":"deepwell-thermal-logo-loop"}]},{"model":"minimax-speech-2-8","air":"minimax:speech@2.8","name":"MiniMax Speech 2.8","creator":"minimax","zdr":true,"status":"live","releasedAt":"2026-01-29T00:00:00Z","weight":4,"headline":"High-quality text-to-speech with expressive, natural voice synthesis","description":"MiniMax Speech 2.8 is an advanced text-to-speech model that turns text into natural, expressive audio in multiple languages. It delivers broadcast-ready speech with rich prosody, emotional control, and a diverse voice library. The model supports up to large input lengths and can be used for voiceovers, narration, accessibility tools, and interactive voice applications.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-speech-2-8.jpg","ogImage":"https://assets.runware.ai/og/minimax-speech-2-8.png","pricingOverview":"Pricing is charged by characters used. HD is $0.10/1,000 characters. Turbo is $0.06/1,000 characters. ","pricingRates":[{"amount":0.0001,"unit":"character","label":"HD"},{"amount":0.00006,"unit":"character","label":"Turbo"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.0094","amount":0.0094,"latencyMs":4083,"exampleId":"fermentation-podcast-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.0239","amount":0.0239,"latencyMs":5065,"exampleId":"lunar-greenhouse-pa-ambience"},{"configuration":"text-to-audio","price":"$0.0442","amount":0.0442,"latencyMs":5628,"exampleId":"community-heat-mapping-explainer-voiceover"},{"configuration":"text-to-audio","price":"$0.065","amount":0.065,"latencyMs":5718,"exampleId":"textile-forensics-podcast-cold-open"}]},{"model":"bfl-flux-2-klein-9b","air":"runware:400@2","name":"FLUX.2 [klein] 9B","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-23T00:00:00Z","weight":4,"headline":"Ultra-fast image generation and editing with sub-second latency","description":"FLUX.2 [klein] 9B is a 4-step distilled image generation and editing model designed for sub-second inference without sacrificing visual quality. It unifies text-to-image and advanced editing workflows in a single model, making it suitable for interactive applications, real-time previews, and latency-critical production use.","architecture":"flux-2-klein-9b","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-klein-9b.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-klein-9b.png","pricingMeasured":[{"configuration":"1024x1024 · 4 steps","price":0.00078}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.00078","amount":0.00078,"latencyMs":1784,"exampleId":"resonator-guitar-diner-background-swap"},{"configuration":"text-to-image · 1024×1024","price":"$0.00169","amount":0.00169,"latencyMs":1789,"exampleId":"municipal-snowplow-ceremony-invitation"},{"configuration":"reference-to-image · 720p","price":"$0.00169","amount":0.00169,"latencyMs":6731,"exampleId":"archaeological-fieldwear-fashion-lookbook"},{"configuration":"reference-to-image · 720p","price":"$0.00338","amount":0.00338,"latencyMs":6091,"exampleId":"community-electronics-repair-social-ad"}]},{"model":"google-gemini-3-flash","air":"google:gemini@3-flash","name":"Gemini 3 Flash","creator":"google","endpoints":["chat-completions"],"status":"live","releasedAt":"2025-12-17T00:00:00Z","weight":4,"headline":"Advanced multimodal text and reasoning model","description":"Gemini 3 Flash is Google’s flagship multimodal language model that processes text alongside images, audio, video, code, and documents. It offers high-performance reasoning, complex instruction following, and deep contextual understanding for a wide range of tasks across language, analysis, and problem solving.","capabilities":["io:text-to-text","io:image-to-text","io:video-to-text","io:audio-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/google-gemini-3-flash.jpg","ogImage":"https://assets.runware.ai/og/google-gemini-3-flash.png","contextWindow":1048576,"maxOutputTokens":65536,"pricingBasis":"token","pricingRates":[{"amount":5e-7,"unit":"inputToken","label":"text, image, video"},{"amount":0.000001,"unit":"inputToken","label":"audio"},{"amount":0.000003,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.00051","amount":0.00051,"latencyMs":12890,"exampleId":"crane-lift-weather-lookup"},{"configuration":"audio-driven","price":"$0.00719","amount":0.007193,"latencyMs":17886,"exampleId":"solar-inverter-service-report"},{"configuration":"reference-to-text","price":"$0.00829","amount":0.008293,"latencyMs":27689,"exampleId":"opposition-corner-kick-scouting-brief"},{"configuration":"text-generation","price":"$0.0105","amount":0.010538,"latencyMs":20608,"exampleId":"hospital-sterilizer-procurement-decision-memo"}]},{"model":"bfl-flux-2-max","air":"bfl:7@1","name":"FLUX.2 [max]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2025-12-16T00:00:00Z","weight":4,"headline":"The latest state-of-the-art model from Black Forest Labs, generating images grounded in live web information.","description":"FLUX.2 [max] is a high-precision text to image and image editing model from Black Forest Labs that generates visuals grounded in real-time information via live web search. It delivers maximum prompt adherence with multi-reference editing and state-of-the-art consistency across identities, objects, and details.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-max.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-max.png","pricingOverview":"Output images cost $0.07 for the first megapixel, then $0.03 per additional megapixel. Reference images cost an additional $0.03 per megapixel.","pricingRates":[{"amount":0.07,"unit":"outputMegapixel","label":"First megapixel out"},{"amount":0.03,"unit":"outputMegapixel","label":"Each further megapixel out","after":1},{"amount":0.03,"unit":"inputMegapixel"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.10","amount":0.1,"latencyMs":36787,"exampleId":"night-pollinator-field-crew-graphic"},{"configuration":"text-to-image · 1536×1024","price":"$0.10","amount":0.1,"latencyMs":34157,"exampleId":"moa-skeleton-habitat-composite"},{"configuration":"reference-to-image · 1024×1536","price":"$0.16","amount":0.16,"latencyMs":71008,"exampleId":"cryogenic-seed-vault-director-portrait"},{"configuration":"reference-to-image · 1024×1536","price":"$0.16","amount":0.16,"latencyMs":56735,"exampleId":"coastal-sound-festival-poster"}]},{"model":"alibaba-wan2-6-image","air":"alibaba:wan@2.6-image","name":"Wan2.6 Image","creator":"alibaba","status":"live","releasedAt":"2025-12-16T00:00:00Z","weight":4,"headline":"High fidelity image generation built on the Wan2.6 visual stack","description":"Wan2.6 Image is a single-frame image generation model derived from the Wan2.6 multimodal video architecture. It focuses on strong prompt adherence, clean spatial structure, and visually coherent results, delivering video-grade image quality for creative, editorial, and product-oriented workflows.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-6-image.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-6-image.png","pricingRates":[{"amount":0.03,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 1024×1600","price":"$0.03","amount":0.03,"latencyMs":16971,"exampleId":"storm-chaser-broadcast-avatar"},{"configuration":"text-to-image · 1600×1024","price":"$0.03","amount":0.03,"latencyMs":16675,"exampleId":"experimental-folk-album-cover"},{"configuration":"reference-to-image · 1024×1600","price":"$0.03","amount":0.03,"latencyMs":12516,"exampleId":"paleobotanist-linocut-profile-avatar"},{"configuration":"text-to-image · 1600×1024","price":"$0.03","amount":0.03,"latencyMs":14913,"exampleId":"conservas-bar-menu-hero-photograph"}]},{"model":"bytedance-seedream-4-5","air":"bytedance:seedream@4.5","name":"Seedream 4.5","creator":"bytedance","status":"live","releasedAt":"2025-12-03T00:00:00Z","weight":4,"headline":"Seedream 4.5 high fidelity multi reference text to image model","description":"Seedream 4.5 is a ByteDance image model for precise 2K to 4K generation and editing. It improves multi image composition, preserves reference detail, and renders small text more reliably. It supports up to 14 reference images for stable characters and design heavy layouts.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-4-5.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-4-5.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1536×2560","price":"$0.04","amount":0.04,"latencyMs":12144,"exampleId":"fermentation-culture-lobby-wall-art"},{"configuration":"text-to-image · 1536×2560","price":"$0.04","amount":0.04,"latencyMs":25121,"exampleId":"mycelium-harvest-supper-invitation"},{"configuration":"reference-to-image · 2560×1536","price":"$0.04","amount":0.04,"latencyMs":43769,"exampleId":"submarine-cable-archive-restoration"},{"configuration":"sequential-images · 4K","price":"$0.04","amount":0.04,"latencyMs":79207,"exampleId":"pinball-factory-archive-colorization"}],"guides":[{"slug":"prompting","title":"Prompting for layered, high-resolution scenes","description":"How to write prompts for Seedream 4.5, from short interpretive prompts to layered scene descriptions, with legible text rendering at 2K to 4K.","date":"2026-06-15","url":"https://runware.ai/docs/models/bytedance-seedream-4-5/guides/prompting"}]},{"model":"bfl-flux-2-pro","air":"bfl:5@1","name":"FLUX.2 [pro]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2025-11-26T00:00:00Z","weight":4,"headline":"High control FLUX.2 Pro image generation and editing","description":"FLUX.2 [pro] is a flow-matching latent transformer for precise text-to-image synthesis and reference-guided editing. It supports multi image references, 4MP outputs, and Mistral-based text conditioning for controllable composition and robust iterative edits that preserve structure.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-pro.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-pro.png","pricingRates":[{"amount":0.03,"unit":"outputMegapixel","label":"First megapixel out"},{"amount":0.015,"unit":"outputMegapixel","label":"Each further megapixel out","after":1},{"amount":0.015,"unit":"inputMegapixel"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.045","amount":0.045,"latencyMs":14348,"exampleId":"aircraft-deicing-operations-blog-hero"},{"configuration":"text-to-image · 1536×1024","price":"$0.045","amount":0.045,"latencyMs":17205,"exampleId":"radio-telescope-farewell-invitation"},{"configuration":"reference-to-image · 1024×1536","price":"$0.075","amount":0.075,"latencyMs":22691,"exampleId":"art-deco-moth-wall-print"},{"configuration":"reference-to-image · 1024×1536","price":"$0.075","amount":0.075,"latencyMs":23424,"exampleId":"offshore-wind-instructor-persona-portrait"}]},{"model":"bfl-flux-2-dev","air":"runware:400@1","name":"FLUX.2 [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-25T00:00:00Z","weight":4,"headline":"FLUX.2 dev for controllable open text to image workflows","description":"FLUX.2 dev is an open weight text to image and image editing model from Black Forest Labs. It targets developers who need precise control over prompts, references, and iteration. Use it for non commercial research, workflow prototyping, and multi conditioning image pipelines.","architecture":"flux-2-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-dev.png","pricingMeasured":[{"configuration":"512×512 · steps 20","price":0.0045,"latencyMs":2265},{"configuration":"1024×1024 · steps 20","price":0.009,"latencyMs":3438},{"configuration":"1024×1024 · steps 50","price":0.0205,"latencyMs":6184}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.009","amount":0.009,"latencyMs":3787,"exampleId":"contemporary-apiarist-workwear-lookbook"},{"configuration":"text-to-image · 1024×1024","price":"$0.009","amount":0.009,"latencyMs":4627,"exampleId":"maritime-artifact-exhibition-scene"},{"configuration":"reference-to-image · 1536×1024","price":"$0.0294","amount":0.0294,"latencyMs":13379,"exampleId":"atacama-observatory-tourism-campaign-hero"},{"configuration":"reference-to-image · 1536×1024","price":"$0.0301","amount":0.0301,"latencyMs":12247,"exampleId":"adaptive-rainwear-transit-lookbook"}]},{"model":"bytedance-seedance-1-0-pro-fast","air":"bytedance:2@2","name":"Seedance 1.0 Pro Fast","creator":"bytedance","status":"live","releasedAt":"2025-10-25T00:00:00Z","weight":4,"headline":"Fast Seedance 1.0 Pro video generation for dance content","description":"Seedance 1.0 Pro Fast accelerates the core Seedance pipeline for expressive dance and performance clips. It turns text prompts or reference images into smooth, cinematic motion with strong temporal consistency. Ideal for rapid iteration in creative tools and production workflows.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-1-0-pro-fast.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-1-0-pro-fast.png","pricingOverview":"≈ $0.0062/s at 480p, $0.0140/s at 720p, and $0.0314/s at 1080p","pricingRates":[{"amount":0.00629,"unit":"durationSecond","label":"16:9 · 480p"},{"amount":0.00629,"unit":"durationSecond","label":"9:16 · 480p"},{"amount":0.00609,"unit":"durationSecond","label":"4:3 · 480p"},{"amount":0.00609,"unit":"durationSecond","label":"3:4 · 480p"},{"amount":0.00623,"unit":"durationSecond","label":"1:1 · 480p"},{"amount":0.00607,"unit":"durationSecond","label":"21:9 · 480p"},{"amount":0.01336,"unit":"durationSecond","label":"16:9 · 720p"},{"amount":0.01336,"unit":"durationSecond","label":"9:16 · 720p"},{"amount":0.01417,"unit":"durationSecond","label":"4:3 · 720p"},{"amount":0.01417,"unit":"durationSecond","label":"3:4 · 720p"},{"amount":0.01402,"unit":"durationSecond","label":"1:1 · 720p"},{"amount":0.01464,"unit":"durationSecond","label":"21:9 · 720p"},{"amount":0.03177,"unit":"durationSecond","label":"16:9 · 1080p"},{"amount":0.03177,"unit":"durationSecond","label":"9:16 · 1080p"},{"amount":0.03159,"unit":"durationSecond","label":"4:3 · 1080p"},{"amount":0.03159,"unit":"durationSecond","label":"3:4 · 1080p"},{"amount":0.03154,"unit":"durationSecond","label":"1:1 · 1080p"},{"amount":0.03072,"unit":"durationSecond","label":"21:9 · 1080p"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 6s","price":"$0.0807","amount":0.08074209,"latencyMs":33106,"exampleId":"sourdough-culture-archive-presenter"},{"configuration":"text-to-video · 720p · 6s","price":"$0.0807","amount":0.08074209,"latencyMs":33991,"exampleId":"aquaculture-workwear-fashion-showcase-reel"},{"configuration":"first-frame · 1080p · 5s","price":"$0.16","amount":0.16019916,"latencyMs":81906,"exampleId":"glassblowing-workshop-social-reel"},{"configuration":"first-frame · 1080p · 5s","price":"$0.16","amount":0.16019916,"latencyMs":41389,"exampleId":"tidal-sensor-network-logo-loop"}]},{"model":"google-veo-3-1","air":"google:3@2","name":"Veo 3.1","creator":"google","zdr":true,"status":"live","releasedAt":"2025-10-15T00:00:00Z","weight":4,"headline":"Veo 3.1 cinematic AI video with native audio","description":"Veo 3.1 is a cinematic video generation model for developers. It turns text prompts or reference images into high fidelity scenes with richer native audio, better prompt adherence, and granular shot control. Use it for story driven clips with smoother motion and consistent style.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/google-veo-3-1.jpg","ogImage":"https://assets.runware.ai/og/google-veo-3-1.png","pricingOverview":"Each generation will cost $0.2/s for 720p, or $0.4/s for 720p with audio.","pricingRates":[{"amount":0.2,"unit":"durationSecond","label":"720p / 1080p · without audio"},{"amount":0.4,"unit":"durationSecond","label":"720p / 1080p · with audio"},{"amount":0.4,"unit":"durationSecond","label":"4K · without audio"},{"amount":0.6,"unit":"durationSecond","label":"4K · with audio"}],"pricingExamples":[{"configuration":"text-to-video · 1080p · 4s · generateAudio","price":"$1.60","amount":1.6,"latencyMs":89836,"exampleId":"ultrasonic-vinyl-cleaning-social-reel"},{"configuration":"first-last-frame · 720p · 6s · generateAudio","price":"$2.40","amount":2.4,"latencyMs":79917,"exampleId":"underwater-camera-housing-colorway-reveal"},{"configuration":"first-frame · 720p · 6s · generateAudio","price":"$2.40","amount":2.4,"latencyMs":57519,"exampleId":"oyster-grading-machine-demo"},{"configuration":"video-edit · 7s","price":"$2.80","amount":2.8,"latencyMs":76391,"exampleId":"arborist-helmet-colorway-reveal"},{"configuration":"reference-to-video · 1080p · 8s · generateAudio","price":"$3.20","amount":3.2,"latencyMs":92297,"exampleId":"antique-automaton-background-replacement"}]},{"model":"google-veo-3-1-fast","air":"google:3@3","name":"Veo 3.1 Fast","creator":"google","zdr":true,"status":"live","releasedAt":"2025-10-15T00:00:00Z","weight":4,"headline":"High speed Google Veo 3.1 Fast text to video generation","description":"Veo 3.1 Fast is a high speed variant of Veo 3.1 for rapid creative iteration. It supports text prompts, image prompts, and reference images. It targets low latency workflows while keeping cinematic quality for short form and multi shot video generation with native audio.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/google-veo-3-1-fast.jpg","ogImage":"https://assets.runware.ai/og/google-veo-3-1-fast.png","pricingOverview":"Each generation will cost $0.1/s for 1080p, or $0.15/s for 1080p with audio.","pricingRates":[{"amount":0.1,"unit":"durationSecond","label":"720p / 1080p"},{"amount":0.15,"unit":"durationSecond","label":"720p / 1080p · with audio"},{"amount":0.3,"unit":"durationSecond","label":"4K"},{"amount":0.35,"unit":"durationSecond","label":"4K · with audio"}],"pricingExamples":[{"configuration":"first-frame · 720p · 4s","price":"$0.60","amount":0.6,"latencyMs":43745,"exampleId":"bronze-bell-foundry-b-roll"},{"configuration":"text-to-video · 1080p · 6s · generateAudio","price":"$0.90","amount":0.9,"latencyMs":81263,"exampleId":"night-transit-director-portrait-loop"},{"configuration":"video-extend · 7s","price":"$1.05","amount":1.05,"latencyMs":80931,"exampleId":"precision-fermentation-sampling-b-roll"},{"configuration":"reference-to-video · 1080p · 8s","price":"$1.20","amount":1.2,"latencyMs":103293,"exampleId":"art-deco-lido-relight-film"},{"configuration":"first-last-frame · 1080p · 8s","price":"$1.20","amount":1.2,"latencyMs":115668,"exampleId":"ammonite-fossil-preparation-b-roll"}]},{"model":"klingai-2-5-turbo-pro","air":"klingai:6@1","name":"KlingAI 2.5 Turbo Pro","creator":"klingai","status":"live","releasedAt":"2025-09-23T00:00:00Z","weight":4,"headline":"Cinematic text to video and image to video at scale","description":"KlingAI 2.5 Turbo Pro is a high performance video generation model for cinematic work. It converts prompts or stills into smooth 1080p clips with strong motion, precise camera control and tight prompt adherence. Ideal for creative tools, ads, trailers and sports scenes.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-2-5-turbo-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-2-5-turbo-pro.png","pricingOverview":"Each generation will cost $0.0700/s for 1080p.","pricingRates":[{"amount":0.07,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"first-last-frame · 5s","price":"$0.35","amount":0.35,"latencyMs":148086,"exampleId":"volcanic-observatory-documentary-titles"},{"configuration":"text-to-video · 5s","price":"$0.35","amount":0.35,"latencyMs":224906,"exampleId":"industrial-sprinkler-activation-explainer-clip"},{"configuration":"text-to-video · 5s","price":"$0.35","amount":0.35,"latencyMs":147042,"exampleId":"mycelium-textile-laboratory-b-roll"},{"configuration":"first-frame · 10s","price":"$0.70","amount":0.7,"latencyMs":69482,"exampleId":"antique-watch-restoration-social-reel"}]},{"model":"lightricks-ltx-2-5-fast","air":"lightricks:ltx@2.5-fast","name":"LTX-2.5 Fast","creator":"lightricks","zdr":true,"status":"live","releasedAt":"2026-08-11T00:00:00Z","weight":3,"headline":"Fast high-resolution video generation with longer clip support, native audio, and first-to-last-frame control","description":"LTX-2.5 Fast is the speed-focused variant in the LTX 2.5 video family, built for high-throughput text-to-video and image-to-video generation. It supports output from 720p through 4K, longer durations at HD resolutions, optional native audio generation, and first-to-last-frame image guidance for more directed motion and shot planning. It is well suited to rapid creative iteration, storyboard development, social and advertising content, and other video workflows that need faster turnaround without dropping to low-end resolutions.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/lightricks-ltx-2-5-fast.jpg","ogImage":"https://assets.runware.ai/og/lightricks-ltx-2-5-fast.png","pricingOverview":"Pricing starts at $0.09 per second for 720p ","pricingRates":[{"amount":0.09,"unit":"durationSecond","label":"720p"},{"amount":0.13,"unit":"durationSecond","label":"1080p"},{"amount":0.19,"unit":"durationSecond","label":"2K"},{"amount":0.3,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"audio-driven · 1080p","price":"$0.78","amount":0.78,"latencyMs":34442,"exampleId":"volcanology-museum-sonic-logo-loop"},{"configuration":"first-last-frame · 1080p · 8s","price":"$1.04","amount":1.04,"latencyMs":53250,"exampleId":"pneumatic-flood-barrier-deployment-demo"},{"configuration":"text-to-video · 720p · 12s · audio","price":"$1.08","amount":1.08,"latencyMs":28528,"exampleId":"rooftop-apiary-honey-harvest-ad"},{"configuration":"first-frame · 1440p · 10s","price":"$1.90","amount":1.9,"latencyMs":63818,"exampleId":"hydrographic-survey-studio-logo-loop"}]},{"model":"lightricks-ltx-2-5-pro","air":"lightricks:ltx@2.5-pro","name":"LTX-2.5 Pro","creator":"lightricks","zdr":true,"status":"live","releasedAt":"2026-08-11T00:00:00Z","weight":3,"headline":"High-fidelity multimodal video generation with native audio, editing workflows, and up to 4K output","description":"LTX-2.5 Pro is the higher-capability model in the LTX 2.5 video family, built for production-quality multimodal video creation and transformation. It supports text-to-video, image-to-video, and audio-to-video workflows together with retake, extension, and reframing operations, pairing synchronized audio generation with higher-resolution output and longer-form clip control. It is well suited to commercial video production, concept films, storyboards, speaking-character shots, and other workflows that need stronger quality, broader modality support, and more flexible post-generation iteration than the Fast tier.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","op:edit","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/lightricks-ltx-2-5-pro.jpg","ogImage":"https://assets.runware.ai/og/lightricks-ltx-2-5-pro.png","pricingOverview":"Pricing starts at $0.12 per second at 720p","pricingRates":[{"amount":0.12,"unit":"durationSecond","label":"720p"},{"amount":0.17,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 6s · audio","price":"$0.72","amount":0.72,"latencyMs":21656,"exampleId":"roller-derby-rink-background-replacement"},{"configuration":"first-frame · 720p · 8s","price":"$0.96","amount":0.96,"latencyMs":25234,"exampleId":"technical-beekeeping-jacket-showcase-reel"},{"configuration":"audio-driven · 1080p","price":"$1.20","amount":1.1971,"latencyMs":30204,"exampleId":"kelp-nursery-safety-presenter"},{"configuration":"first-last-frame · 1080p · 10s","price":"$1.70","amount":1.7,"latencyMs":40195,"exampleId":"courtroom-sketch-artist-headshot-retouch"}],"guides":[{"slug":"prompting","title":"Prompting","description":"How to write text-to-video prompts for LTX-2.5 Pro: the six-part shot scaffold, directing the action and camera, matching detail to shot scale, and prompting its native audio.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/prompting"},{"slug":"camera-movement","title":"Camera movement","description":"How to control the camera in LTX-2.5 Pro: the eight settings.cameraMovement presets, the prompt camera vocabulary, and when to reach for a reliable preset versus a described move.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/camera-movement"},{"slug":"multi-shot","title":"Multi-shot sequences","description":"How to prompt LTX-2.5 Pro for multi-shot video: cutting between shots in one generation, naming transitions, holding character and audio across cuts, and giving each shot a job.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/multi-shot"},{"slug":"native-audio","title":"Native audio","description":"How to generate synchronized audio with LTX-2.5 Pro: turning on settings.audio, prompting ambient sound and effects, directing spoken dialogue with accent and lip-sync, and balancing the mix.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/native-audio"},{"slug":"first-and-last-frame","title":"First and last frame","description":"How to animate images with LTX-2.5 Pro using inputs.frameImages: turning a still into motion, directing the ending with a last frame, and building clean loops.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/first-and-last-frame"},{"slug":"audio-driven","title":"Audio-driven characters","description":"How to drive a lip-synced performance with LTX-2.5 Pro: pairing inputs.audio with a reference image, matching voice to subject, and the length, resolution, and framing rules.","date":"2026-08-11","url":"https://runware.ai/docs/models/lightricks-ltx-2-5-pro/guides/audio-driven"}]},{"model":"luma-ray3-2","air":"luma:ray@3.2","name":"Ray3.2","creator":"luma","status":"live","releasedAt":"2026-06-09T00:00:00Z","weight":3,"headline":"Cinematic video model for generation, transformation, motion transfer, and frame-level direction","description":"Ray3.2 is Luma's flagship video model for turning creative direction into controllable production workflows. It supports text-to-video, image-to-video, and video-to-video generation, with stronger continuity, motion transfer, camera motion transfer, character transformation, relighting, environment change, and product-swap workflows. It is built for cinematic-quality output, multi-keyframe control inside a single clip, and Modify Video V2 workflows that preserve performance, lighting, and scene structure while transforming existing footage.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/luma-ray3-2.jpg","ogImage":"https://assets.runware.ai/og/luma-ray3-2.png","pricingOverview":"Video generation and Video Edit have different pricing.\nVideo generation:\nPricing is highly affected by resolution starting at $0.06 for 5s at 360p. Duration available are 5s and 10s. \nHDR and HDR + EXR are only for 5s and increase costs ($0.60/$0.90 at 720p and $2.40/$3.60 at 1080p).\nExtend is standard dynamic range only. It always bills as one 5-second block, even if video.duration is present on the request. it costs $0.15 at 540p, $0.30 at 720p and $1.20 at 1080p.\nReframe is standard dynamic range only and bills per second. $0.03/s at 320p, $0.06/s at 540p, $0.12/s at 720p, and $0.36/s at 1080p.\n\nVideo Edit:\nVideo edits use their own rate table. HDR, and HDR + EXR edits increase costs. Each have separate 5-second and 10-second totals. HDR: from $1.08 (360p) to $4.32 (1080p) for 5s. HDR+EXR: from $1.62 (360p) to $6.48 (1080p) for 5s\n","pricingRates":[{"amount":0.012,"unit":"durationSecond","label":"Text/image to video · 360p"},{"amount":0.03,"unit":"durationSecond","label":"Text/image to video · 540p"},{"amount":0.06,"unit":"durationSecond","label":"Text/image to video · 720p"},{"amount":0.24,"unit":"durationSecond","label":"Text/image to video · 1080p"},{"amount":0.12,"unit":"durationSecond","label":"HDR · 720p"},{"amount":0.48,"unit":"durationSecond","label":"HDR · 1080p"},{"amount":0.18,"unit":"durationSecond","label":"HDR+EXR · 720p"},{"amount":0.72,"unit":"durationSecond","label":"HDR+EXR · 1080p"},{"amount":0.108,"unit":"durationSecond","label":"Video edit · 360p"},{"amount":0.144,"unit":"durationSecond","label":"Video edit · 540p"},{"amount":0.216,"unit":"durationSecond","label":"Video edit · 720p"},{"amount":0.432,"unit":"durationSecond","label":"Video edit · 1080p"},{"amount":0.15,"unit":"output","label":"Extend · 540p"},{"amount":0.3,"unit":"output","label":"Extend · 720p"},{"amount":1.2,"unit":"output","label":"Extend · 1080p"},{"amount":0.03,"unit":"durationSecond","label":"Reframe · 320p"},{"amount":0.06,"unit":"durationSecond","label":"Reframe · 540p"},{"amount":0.12,"unit":"durationSecond","label":"Reframe · 720p"},{"amount":0.36,"unit":"durationSecond","label":"Reframe · 1080p"},{"amount":0.432,"unit":"durationSecond","label":"Video edit HDR · 720p"},{"amount":0.648,"unit":"durationSecond","label":"Video edit HDR+EXR · 720p"},{"amount":0.864,"unit":"durationSecond","label":"Video edit HDR · 1080p"},{"amount":1.296,"unit":"durationSecond","label":"Video edit HDR+EXR · 1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 540p · 5s","price":"$0.15","amount":0.15,"latencyMs":36984,"exampleId":"geothermal-network-logo-sting"},{"configuration":"multiple-keyframes · 720p · 5s","price":"$0.30","amount":0.3,"latencyMs":61651,"exampleId":"artisan-paper-marbling-b-roll"},{"configuration":"video-reframe · 720p","price":"$0.725","amount":0.7250394,"latencyMs":83166,"exampleId":"planetarium-archive-frame-restoration"},{"configuration":"text-to-video · 720p · 5s","price":"$0.90","amount":0.9,"latencyMs":86663,"exampleId":"colorized-apiary-newsreel-restoration"},{"configuration":"video-edit · 720p","price":"$1.08","amount":1.08,"latencyMs":193595,"exampleId":"veterinary-ultrasound-field-demo"},{"configuration":"first-frame · 1080p · 5s","price":"$1.20","amount":1.2,"latencyMs":509020,"exampleId":"chamber-orchestra-group-portrait-clip"}],"guides":[{"slug":"generating-video","title":"Generating video from text and images","description":"How to generate cinematic video with Luma Ray 3.2: text-to-video, image-to-video, frame-level keyframes, and the resolution, duration, HDR, and loop controls.","date":"2026-06-09","url":"https://runware.ai/docs/models/luma-ray3-2/guides/generating-video"},{"slug":"editing-video","title":"Transforming and restyling video","description":"How to transform footage with Luma Ray 3.2: restyle or reskin a clip while its motion carries through, using the strength dial and per-signal conditioning controls.","date":"2026-06-09","url":"https://runware.ai/docs/models/luma-ray3-2/guides/editing-video"},{"slug":"reframing-video","title":"Reframing video for any aspect ratio","description":"How to reframe footage with Luma Ray 3.2: convert a clip to a new aspect ratio or a larger canvas while the model extends the scene, using inputs.video and sourcePosition.","date":"2026-06-09","url":"https://runware.ai/docs/models/luma-ray3-2/guides/reframing-video"}]},{"model":"klingai-video-3-0-omni-4k","air":"klingai:kling-video@o3-4k","name":"Kling VIDEO 3.0 Omni 4K","creator":"klingai","status":"live","releasedAt":"2026-04-23T00:00:00Z","weight":3,"headline":"4K unified multimodal video generation with native audio, editing, and higher-end narrative control","description":"Kling VIDEO 3.0 Omni 4K is the 4K variant of Kling VIDEO 3.0 Omni for text-to-video and image-to-video workflows. It raises the 3.0 Omni line from 720p Standard and 1080p Pro to 4K output while preserving the series strengths: native audio generation, reference-guided video creation, prompt-based editing, multi-shot structure, and stable subject consistency for more demanding cinematic and advertising workflows.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-o3-4k.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-o3-4k.png","pricingRates":[{"amount":0.42,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"first-frame · 5s","price":"$2.10","amount":2.1,"latencyMs":102041,"exampleId":"aquaculture-sensor-cleaning-demo"},{"configuration":"video-edit · 4k","price":"$2.52","amount":2.52,"latencyMs":191057,"exampleId":"bauhaus-acoustic-panel-social-reel"},{"configuration":"text-to-video · 2160p · 6s","price":"$2.52","amount":2.52,"latencyMs":108931,"exampleId":"fountain-pen-restoration-social-reel"},{"configuration":"reference-to-video · 2160p · 7s","price":"$2.94","amount":2.94,"latencyMs":205452,"exampleId":"urban-bat-ecologist-update"},{"configuration":"first-last-frame · 8s","price":"$3.36","amount":3.36,"latencyMs":165484,"exampleId":"portable-manuscript-scanner-explainer"},{"configuration":"multi-shot · 2160p · 9s","price":"$3.78","amount":3.78,"latencyMs":436067,"exampleId":"cathedral-pipe-organ-maintenance-b-roll"},{"configuration":"elements · 2880×2880 · 10s","price":"$4.20","amount":4.2,"latencyMs":284581,"exampleId":"mushroom-forager-workwear-fashion-reel"}]},{"model":"sync-3","air":"sync:3@0","name":"sync-3","creator":"sync","status":"live","releasedAt":"2026-04-07T00:00:00Z","weight":3,"headline":"Full-scene lip synchronization with global face understanding and obstruction handling","description":"sync-3 is a lip synchronization model that processes entire shots as a single generation rather than stitching independent segments. It builds a global understanding of the speaker across all frames, enabling consistent output on close-ups, extreme face angles, partially obscured faces, and obstructed mouths. The model preserves the original speaker's style, cadence, and emotional expression across 95+ languages.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sync-3.jpg","ogImage":"https://assets.runware.ai/og/sync-3.png","pricingRates":[{"amount":0.133,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"segmented-lip-sync","price":"$0.804","amount":0.8035461,"latencyMs":121320,"exampleId":"maritime-documentary-series-teaser"},{"configuration":"audio-driven","price":"$1.06","amount":1.064,"latencyMs":153214,"exampleId":"seed-vault-conspiracy-teaser"},{"configuration":"audio-driven","price":"$1.11","amount":1.113875,"latencyMs":80540,"exampleId":"mobile-equine-dentistry-service-reel"}]},{"model":"exactly-illustrative-editorial-line","air":"exactly:illustrative@editorial-line","name":"Exactly Editorial Line","creator":"exactly","status":"live","releasedAt":"2026-04-02T00:00:00Z","weight":3,"headline":"Clean, editorial-style line illustrations with refined detail","description":"Editorial Line is an illustrative style from Exactly AI that produces clean, refined line-based illustrations with an editorial sensibility. It generates artwork with precise linework and measured compositions, ideal for publishing, magazine visuals, storytelling, and professional editorial content.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-editorial-line.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-editorial-line.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"prunaai-p-image-upscale","air":"prunaai:p-image@upscale","name":"P-Image Upscale","creator":"prunaai","status":"live","releasedAt":"2026-04-02T00:00:00Z","weight":3,"headline":"AI-powered image upscaling up to 8 megapixels with detail and realism enhancement","description":"P-Image Upscale is an image upscaling model from Pruna that increases resolution using AI with a configurable target from 1 to 8 megapixels. It includes optional detail enhancement and realism enhancement controls for improving output quality, particularly effective on AI-generated images. It supports JPEG and PNG output formats with both synchronous and asynchronous processing.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-image-upscale.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-image-upscale.png","pricingOverview":"Pricing vary based on target resolution: 1–4 MP is $0.005 / image and 5–8 MP is $0.01 / image","pricingRates":[{"amount":0.005,"unit":"output","label":"1–4 MP"},{"amount":0.01,"unit":"output","label":"5–8 MP"},{"amount":0.02,"unit":"output","label":"9-16 MP"},{"amount":0.04,"unit":"output","label":"17-32 MP"},{"amount":0.06,"unit":"output","label":"33-64 MP"},{"amount":0.12,"unit":"output","label":"65-128 MP"}],"pricingExamples":[{"configuration":"upscale","price":"$0.02","amount":0.02,"latencyMs":14431,"exampleId":"deep-sea-exhibit-wall-panorama"},{"configuration":"upscale","price":"$0.02","amount":0.02,"latencyMs":19384,"exampleId":"lichen-green-microscope-colorway-master"},{"configuration":"upscale","price":"$0.02","amount":0.02,"latencyMs":16256,"exampleId":"falconry-hood-auction-catalog-image"},{"configuration":"upscale","price":"$0.02","amount":0.02,"latencyMs":26488,"exampleId":"historic-glasshouse-cleanup-master"}]},{"model":"ideogram-layerize-text","air":"ideogram:layerize-text@0","name":"Ideogram Layerize Text","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2026-04-01T00:00:00Z","weight":3,"headline":"Extract editable text layers and a text-erased base image from design visuals","description":"Ideogram Layerize Text analyzes an image, detects readable text regions, and turns each line into structured editable text blocks with position, styling, and font information. It also returns a text-erased base image, making it useful for posters, social graphics, packaging, book covers, and other design workflows where copy needs to change without regenerating the whole composition.","capabilities":["io:image-to-image","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-layerize-text.jpg","ogImage":"https://assets.runware.ai/og/ideogram-layerize-text.png","pricingRates":[{"amount":0.09,"unit":"output"}],"pricingExamples":[{"configuration":"text-layer-extraction","price":"$0.09","amount":0.09,"latencyMs":43452,"exampleId":"roller-derby-archive-photo-restoration"},{"configuration":"text-layer-extraction","price":"$0.09","amount":0.09,"latencyMs":25664,"exampleId":"barn-owl-release-event-poster"},{"configuration":"text-layer-extraction","price":"$0.09","amount":0.09,"latencyMs":35639,"exampleId":"arctic-negative-restoration-exhibit"},{"configuration":"text-layer-extraction","price":"$0.09","amount":0.09,"latencyMs":27511,"exampleId":"apiary-jacket-try-on-campaign"}]},{"model":"exactly-photo-bright-pulse","air":"exactly:photo@bright-pulse","name":"Exactly Bright Pulse","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Bright, energetic photographic style with vivid lighting","description":"Bright Pulse is a photographic style from Exactly AI that produces images with vivid, energetic lighting and a fresh, modern feel. It generates photos with bright tones, clean highlights, and lively color balance, ideal for product photography, lifestyle content, social media visuals, and projects that need a vibrant, upbeat aesthetic.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-bright-pulse.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-bright-pulse.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":177848,"exampleId":"geothermal-annual-report-portrait-restyle"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":138093,"exampleId":"pipe-organ-editorial-restyle"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":117498,"exampleId":"satellite-assembly-campaign-hero"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":133468,"exampleId":"airport-falconry-crew-group-portrait"}]},{"model":"exactly-illustrative-dark-comics","air":"exactly:illustrative@dark-comics","name":"Exactly Dark Comics","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Dark, gritty comic art style with heavy shadows and noir aesthetics","description":"Dark Comics is an illustrative style from Exactly AI that produces gritty, noir-inspired comic artwork with heavy shadows and dramatic contrast. It generates illustrations with a dark, moody atmosphere and bold compositions, ideal for horror, thriller, dystopian themes, and projects that demand an intense, cinematic comic book aesthetic.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-dark-comics.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-dark-comics.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-photo-distant-reality","air":"exactly:photo@distant-reality","name":"Exactly Distant Reality","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Dreamy photographic style with surreal, distant atmosphere","description":"Distant Reality is a photographic style from Exactly AI that produces images with a dreamy, surreal quality and an otherworldly atmosphere. It generates photos with soft focus, ethereal lighting, and a sense of detachment from the everyday, suited for conceptual photography, mood boards, and creative projects that aim for an abstract, evocative visual tone.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-distant-reality.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-distant-reality.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":123910,"exampleId":"air-ambulance-pilot-speaker-headshot"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":106374,"exampleId":"kintsugi-conservator-fellowship-headshot"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":126204,"exampleId":"maritime-museum-campaign-hero"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":203173,"exampleId":"planetarium-anniversary-campaign-artwork"}]},{"model":"exactly-illustrative-earthy-elegance","air":"exactly:illustrative@earthy-elegance","name":"Exactly Earthy Elegance","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Warm, organic illustrative style with muted earth tones","description":"Earthy Elegance is an illustrative style from Exactly AI that produces artwork with warm, organic earth tones and a refined aesthetic. It generates illustrations with muted natural palettes and elegant compositions, suited for editorial work, nature-themed visuals, and designs that call for a grounded, sophisticated feel.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-earthy-elegance.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-earthy-elegance.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-photo-extreme-contrast","air":"exactly:photo@extreme-contrast","name":"Exactly Extreme Contrast","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"High-contrast photographic style with dramatic light and shadow","description":"Extreme Contrast is a photographic style from Exactly AI that produces images with dramatic, high-contrast lighting and deep shadow detail. It generates photos with bold tonal separation and strong visual impact, well suited for portrait photography, dramatic compositions, fashion visuals, and any project that demands a powerful chiaroscuro effect.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-extreme-contrast.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-extreme-contrast.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":124540,"exampleId":"museum-conservation-team-composite-portrait"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":175311,"exampleId":"film-archivist-press-kit-headshot"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":128152,"exampleId":"apiary-smoker-field-product-hero"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":134615,"exampleId":"master-perfumer-conference-speaker-headshot"}]},{"model":"exactly-photo-grain-film-look","air":"exactly:photo@grain-film-look","name":"Exactly Grain Film Look","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Analog film photography style with natural grain and warm tones","description":"Grain Film Look is a photographic style from Exactly AI that emulates the look of analog film photography. It generates images with natural film grain, warm color shifts, and the organic imperfections characteristic of 35mm and medium format film, ideal for editorial photography, vintage aesthetics, and projects that benefit from an authentic, nostalgic film quality.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-grain-film-look.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-grain-film-look.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":143507,"exampleId":"night-bus-dispatcher-editorial-headshot"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":109705,"exampleId":"geothermal-field-report-opener"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":197666,"exampleId":"rare-book-conservator-headshot"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":119758,"exampleId":"racing-wheelchair-trackside-background-replacement"}]},{"model":"exactly-illustrative-graphic-harmony","air":"exactly:illustrative@graphic-harmony","name":"Exactly Graphic Harmony","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Balanced, harmonious graphic illustrations with cohesive composition","description":"Graphic Harmony is an illustrative style from Exactly AI that produces balanced, visually cohesive graphic artwork. It generates illustrations with harmonious color relationships and well-structured compositions, well suited for branding, marketing materials, and design projects that require visual consistency.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-graphic-harmony.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-graphic-harmony.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-illustrative-graphic-novel","air":"exactly:illustrative@graphic-novel","name":"Exactly Graphic Novel","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Comic book and graphic novel style with strong ink lines and dramatic shading","description":"Graphic Novel is an illustrative style from Exactly AI that produces artwork in the tradition of graphic novels and comic books. It generates illustrations with strong ink lines, dramatic shading, and panel-ready compositions, well suited for storytelling, sequential art, comic covers, and projects that call for a bold, narrative-driven visual style.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-graphic-novel.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-graphic-novel.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-illustrative-graphite-creature","air":"exactly:illustrative@graphite-creature","name":"Exactly Graphite Creature","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Textured graphite-style illustrations with creature and character focus","description":"Graphite Creature is an illustrative style from Exactly AI that produces textured, pencil-like artwork with a focus on creature and character design. It generates illustrations with rich graphite shading and organic detail, ideal for concept art, fantasy creatures, character studies, and projects that benefit from a hand-drawn, sketchbook aesthetic.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-graphite-creature.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-graphite-creature.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-photo-journey","air":"exactly:photo@journey","name":"Exactly Journey","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Travel and adventure photographic style with rich, cinematic tones","description":"Journey is a photographic style from Exactly AI that produces images with rich, cinematic tones evoking travel and exploration. It generates photos with warm golden-hour lighting, expansive compositions, and a sense of narrative depth, ideal for travel content, landscape photography, campaign visuals, and storytelling projects with an adventurous spirit.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-journey.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-journey.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":136674,"exampleId":"atacama-observatory-sales-composite"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":175061,"exampleId":"1930s-river-mailboat-photo-restoration"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":123969,"exampleId":"peregrine-release-campaign-hero"},{"configuration":"reference-to-image","price":"$0.195","amount":0.195,"latencyMs":212639,"exampleId":"volcanologist-expedition-report-hero"}]},{"model":"exactly-illustrative-monochrome-cafe","air":"exactly:illustrative@monochrome-cafe","name":"Exactly Monochrome Café","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Monochromatic illustrative style with warm café-inspired tones","description":"Monochrome Café is an illustrative style from Exactly AI that produces artwork in warm monochromatic palettes reminiscent of café sketches and ink drawings. It generates illustrations with a cozy, hand-drawn quality using limited tonal ranges, ideal for lifestyle visuals, storyboard art, and projects with a warm, intimate feel.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-monochrome-cafe.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-monochrome-cafe.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-illustrative-muted-modern","air":"exactly:illustrative@muted-modern","name":"Exactly Muted Modern","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Contemporary illustrative style with soft, muted color palettes","description":"Muted Modern is an illustrative style from Exactly AI that produces contemporary artwork with soft, desaturated color palettes and a modern aesthetic. It generates illustrations with understated tones and clean design sensibilities, ideal for modern branding, app visuals, and projects that favor a minimalist, polished look.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-muted-modern.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-muted-modern.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-illustrative-playful-line-adventures","air":"exactly:illustrative@playful-line-adventures","name":"Exactly Playful Line Adventures","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Whimsical, playful line art with an adventurous character","description":"Playful Line Adventures is an illustrative style from Exactly AI that produces whimsical, energetic line art with a sense of adventure and fun. It generates illustrations with loose, expressive linework and a lighthearted feel, perfect for children's content, character design, playful branding, and projects that need a sense of joy and movement.","architecture":"exactly-illustrative","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-illustrative-playful-line-adventures.jpg","ogImage":"https://assets.runware.ai/og/exactly-illustrative-playful-line-adventures.png","pricingOverview":"Inference ready-made 1K illustration styles","pricingRates":[{"amount":0.195,"unit":"output"}]},{"model":"exactly-photo-warm-light","air":"exactly:photo@warm-light","name":"Exactly Warm Light","creator":"exactly","status":"live","releasedAt":"2026-03-30T00:00:00Z","weight":3,"headline":"Soft, warm-lit photographic style with inviting golden tones","description":"Warm Light is a photographic style from Exactly AI that produces images bathed in soft, golden lighting with an inviting warmth. It generates photos with gentle highlights, amber tones, and a cozy atmosphere, perfect for portrait photography, interior visuals, food photography, and lifestyle content that benefits from a naturally warm and welcoming feel.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/exactly-photo-warm-light.jpg","ogImage":"https://assets.runware.ai/og/exactly-photo-warm-light.png","pricingOverview":"Inference photography styles at $0.195 in 1K or 2K and $0.338 in 4K","pricingRates":[{"amount":0.195,"unit":"output","label":"1K"},{"amount":0.195,"unit":"output","label":"2K"},{"amount":0.338,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"relight","price":"$0.195","amount":0.195,"latencyMs":171041,"exampleId":"geothermal-fieldwork-magazine-opener"},{"configuration":"relight","price":"$0.195","amount":0.195,"latencyMs":83305,"exampleId":"former-quarry-solar-farm-relight"},{"configuration":"relight","price":"$0.195","amount":0.195,"latencyMs":97877,"exampleId":"maritime-museum-campaign-portrait"},{"configuration":"relight","price":"$0.195","amount":0.195,"latencyMs":92849,"exampleId":"lavender-field-microscope-colorway"}]},{"model":"topazlabs-starlight-precise-2-5","air":"topazlabs:starlight-precise@2.5","name":"Topaz Labs Starlight Precise 2.5","creator":"topazlabs","zdr":true,"status":"live","releasedAt":"2026-03-25T00:00:00Z","weight":3,"headline":"Diffusion-based video enhancement with photorealistic detail and temporal consistency","description":"Starlight Precise 2.5 is a diffusion-based video enhancement model from Topaz Labs that upscales, denoises, de-aliases, and sharpens video while maintaining full temporal consistency across frames. It delivers photorealistic detail with natural skin textures and faces, preserving the original look of the footage without artificial artifacts.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/topazlabs-starlight-precise-2-5.jpg","ogImage":"https://assets.runware.ai/og/topazlabs-starlight-precise-2-5.png","pricingOverview":"1K upscaling - 24 frames/second costs $0.080 per second\n4K upscaling - 24 frames/second costs $0.175 per second\n","pricingRates":[{"amount":0.08,"unit":"durationSecond","label":"1K"},{"amount":0.175,"unit":"durationSecond","label":"4K"}],"pricingExamples":[{"configuration":"upscale","price":"$0.485","amount":0.4848,"latencyMs":599808,"exampleId":"handmade-paper-music-video-segment"},{"configuration":"upscale","price":"$0.485","amount":0.4848,"latencyMs":170097,"exampleId":"maritime-artifact-conservation-b-roll"},{"configuration":"upscale","price":"$0.485","amount":0.4848,"latencyMs":252968,"exampleId":"artisan-knife-sharpening-social-reel"},{"configuration":"upscale","price":"$0.485","amount":0.4848,"latencyMs":188770,"exampleId":"honey-refractometer-product-demo"}]},{"model":"minimax-m2-7-highspeed","air":"minimax:m2.7@highspeed","name":"MiniMax M2.7 Highspeed","creator":"minimax","endpoints":["chat-completions"],"status":"deprecated","replacedBy":"minimax-m3","deactivatesAt":"2026-11-18T00:00:00Z","releasedAt":"2026-03-18T00:00:00Z","weight":3,"headline":"Faster throughput for agentic coding and tool-driven automation","description":"MiniMax M2.7-Highspeed is the performance-tuned variant of M2.7, built for lower latency and higher throughput while keeping output behavior consistent with the standard model. It’s a strong fit for interactive coding agents, tool-calling pipelines, and office automation flows where responsiveness matters.","capabilities":["io:text-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-m2-7-highspeed.jpg","ogImage":"https://assets.runware.ai/og/minimax-m2-7-highspeed.png","contextWindow":204800,"maxOutputTokens":131072,"pricingBasis":"token","pricingRates":[{"amount":6e-7,"unit":"inputToken"},{"amount":0.0000024,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000358","amount":0.000358,"latencyMs":2102,"exampleId":"concert-ticketing-status-gate"},{"configuration":"tool-calling","price":"$0.000376","amount":0.000376,"latencyMs":2093,"exampleId":"saas-incident-status-reply"},{"configuration":"text-generation","price":"$0.00926","amount":0.009259,"latencyMs":82219,"exampleId":"museum-exhibition-audio-guide-script"},{"configuration":"text-generation","price":"$0.0228","amount":0.022817,"latencyMs":222439,"exampleId":"checkout-accessibility-remediation-backlog"}]},{"model":"openai-gpt-5-4-mini","air":"openai:gpt@5.4-mini","name":"GPT-5.4 Mini","creator":"openai","endpoints":["chat-completions","responses"],"status":"live","releasedAt":"2026-03-17T00:00:00Z","weight":3,"headline":"Efficient reasoning LLM with 400K context for coding assistants and subagent workflows","description":"GPT-5.4 Mini is a compact, efficient variant of GPT-5.4 designed for coding assistants, subagent orchestration, and multimodal applications requiring faster responsiveness. It supports a 400K token context window and retains native computer use and configurable reasoning effort at a lower cost than the flagship model.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-4-mini.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-4-mini.png","contextWindow":400000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":7.5e-7,"unit":"inputToken"},{"amount":0.0000045,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000224","amount":0.000224,"latencyMs":1183,"exampleId":"production-deployment-timing-brief"},{"configuration":"search-grounded","price":"$0.0103","amount":0.010333,"latencyMs":7692,"exampleId":"eu-ai-act-compliance-update"},{"configuration":"code-interpreter","price":"$0.0128","amount":0.012842,"latencyMs":15791,"exampleId":"neighborhood-bistro-menu-profitability-analysis"},{"configuration":"text-generation","price":"$0.0273","amount":0.027281,"latencyMs":46310,"exampleId":"subscription-billing-migration-decision-memo"},{"configuration":"reference-to-text","price":"$0.0308","amount":0.03083,"latencyMs":60325,"exampleId":"roof-damage-triage-report"}]},{"model":"openai-gpt-5-4-nano","air":"openai:gpt@5.4-nano","name":"GPT-5.4 Nano","creator":"openai","endpoints":["chat-completions","responses"],"status":"deprecated","replacedBy":"openai-gpt-6-luna","deactivatesAt":"2027-04-01T00:00:00Z","releasedAt":"2026-03-17T00:00:00Z","weight":3,"headline":"Ultra-low-latency LLM for high-volume classification, extraction, and lightweight automation","description":"GPT-5.4 Nano is the smallest and fastest variant of GPT-5.4, designed for high-throughput, low-latency tasks such as classification, data extraction, ranking, and lightweight automation. It prioritizes speed and cost efficiency for simple, high-volume workloads and is available exclusively via the API.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-4-nano.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-4-nano.png","contextWindow":400000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":2e-7,"unit":"inputToken"},{"amount":0.00000125,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000056","amount":0.000056,"latencyMs":1182,"exampleId":"pharmacovigilance-sla-time-check"},{"configuration":"reference-to-text","price":"$0.00103","amount":0.001034,"latencyMs":7569,"exampleId":"secondhand-jacket-listing-record"},{"configuration":"reference-to-text","price":"$0.00143","amount":0.001433,"latencyMs":7573,"exampleId":"apartment-move-out-damage-report"},{"configuration":"text-generation","price":"$0.00234","amount":0.002343,"latencyMs":12671,"exampleId":"enterprise-renewal-risk-memo"}]},{"model":"xai-tts","air":"xai:tts@0","name":"xAI Text-to-Speech","creator":"xai","zdr":true,"status":"live","releasedAt":"2026-03-16T00:00:00Z","weight":3,"headline":"Expressive text-to-speech with over two dozen voices, speech tags, and multilingual support","description":"xAI Text-to-Speech converts text into natural-sounding spoken audio with a single API call. It offers more than two dozen expressive voices, inline speech tags for fine-grained control over pauses, laughter, whispers, and emphasis, and supports over 20 auto-detected languages.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/xai-tts.jpg","ogImage":"https://assets.runware.ai/og/xai-tts.png","pricingOverview":"$0.015 per 1000 characters","pricingRates":[{"amount":0.000015,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.00583","amount":0.005835,"latencyMs":5180,"exampleId":"antarctic-research-station-pa-announcement"},{"configuration":"text-to-audio","price":"$0.00612","amount":0.00612,"latencyMs":5168,"exampleId":"forensic-arboriculture-podcast-intro"},{"configuration":"text-to-audio","price":"$0.00612","amount":0.00612,"latencyMs":5081,"exampleId":"grid-resilience-podcast-cold-open"},{"configuration":"text-to-audio","price":"$0.00791","amount":0.007905,"latencyMs":5862,"exampleId":"restaurant-oyster-shell-recycling-explainer"}]},{"model":"openai-gpt-5-4-pro","air":"openai:gpt@5.4-pro","name":"GPT-5.4 Pro","creator":"openai","endpoints":["chat-completions"],"status":"live","releasedAt":"2026-03-05T00:00:00Z","weight":3,"headline":"Enterprise-grade reasoning LLM optimized for high-performance professional workloads","description":"GPT-5.4 Pro is the high-performance variant of GPT-5.4, optimized for enterprise-grade professional tasks. It offers deeper reasoning, enhanced accuracy, and extended compute for complex multi-step workflows including document creation, spreadsheet analysis, and autonomous agent orchestration. It shares the 1 million token context window and native computer use capabilities of the standard GPT-5.4.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-4-pro.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-4-pro.png","contextWindow":1050000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":0.00003,"unit":"inputToken"},{"amount":0.00018,"unit":"outputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.0341","amount":0.03414,"latencyMs":20212,"exampleId":"incident-timestamp-tool-invocation"},{"configuration":"search-grounded","price":"$4.82","amount":4.82169,"latencyMs":375550,"exampleId":"fomc-refinancing-impact-brief"},{"configuration":"reference-to-text","price":"$7.57","amount":7.56843,"latencyMs":530768,"exampleId":"retail-shelf-compliance-audit"},{"configuration":"text-generation","price":"$8.37","amount":8.37213,"latencyMs":935254,"exampleId":"rural-maternity-service-decision-memo"}]},{"model":"alibaba-z-image-turbo-controlnet","air":"runware:z-image@turbo-controlnet","name":"Z-Image-Turbo ControlNet Union","creator":"alibaba","status":"live","releasedAt":"2026-02-26T00:00:00Z","weight":3,"headline":"Unified Z-Image Turbo ControlNet for multi-conditioning across canny, depth, pose, MLSD, hed, scribble, gray, and inpaint workflows","description":"Z-Image Turbo ControlNet (Alibaba PAI) is the unified multi-conditioning ControlNet for the Z-Image Turbo family. It supports a broad range of structural control signals including canny, depth, pose, MLSD, hed, scribble, gray, and inpaint-style workflows, making it useful when one ControlNet checkpoint needs to cover several conditioning types inside the same image pipeline.","architecture":"z-image-turbo","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/alibaba-z-image-turbo-controlnet.jpg","ogImage":"https://assets.runware.ai/og/alibaba-z-image-turbo-controlnet.png"},{"model":"recraft-v4","air":"recraft:v4@0","name":"Recraft V4","creator":"recraft","status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":3,"headline":"Professional text-to-image model for brand and marketing design","description":"Recraft V4 is a professional-grade text-to-image model built for design and marketing workflows. It focuses on refined visual aesthetics, strong photorealism, and reliable brand control. The model delivers realistic skin rendering, natural textures, distinctive lighting, and well-structured compositions while avoiding common synthetic artifacts. It supports 2K image generation, color palette control, and explicit background color selection to help teams produce brand-consistent creative assets.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":18080,"exampleId":"cubesat-deorbit-explainer-infographic"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":13072,"exampleId":"urban-infrastructure-magazine-cover"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":16051,"exampleId":"subsea-cable-repair-video-thumbnail"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":14045,"exampleId":"technical-apiary-workwear-lookbook"}]},{"model":"recraft-v4-pro","air":"recraft:v4-pro@0","name":"Recraft V4 Pro","creator":"recraft","status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":3,"headline":"Advanced design-focused image generation with enhanced control and fidelity","description":"Recraft V4 Pro is an advanced text-to-image model tailored for high-end creative production and brand-critical design work. It delivers elevated photorealism, nuanced lighting, refined composition, and contemporary styling suited for professional campaigns. The model provides enhanced control over color palettes, background colors, and style references, enabling precise brand alignment at 2K resolution. It is built to produce distinctive visuals with consistent aesthetic quality across marketing, advertising, and product-focused content.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-pro.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-pro.png","pricingRates":[{"amount":0.25,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.25","amount":0.25,"latencyMs":15054,"exampleId":"urban-bat-roost-retrofit-illustration"},{"configuration":"text-to-image","price":"$0.25","amount":0.25,"latencyMs":17519,"exampleId":"oyster-farm-salinity-meter-campaign"},{"configuration":"text-to-image","price":"$0.25","amount":0.25,"latencyMs":16681,"exampleId":"urban-archaeology-magazine-cover"},{"configuration":"text-to-image","price":"$0.25","amount":0.25,"latencyMs":18804,"exampleId":"semiconductor-fab-manga-cover-art"}]},{"model":"recraft-v4-pro-vector","air":"recraft:v4-pro@vector","name":"Recraft V4 Pro Vector","creator":"recraft","status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":3,"headline":"Advanced vectorization with enhanced control for professional workflows","description":"Recraft V4 Pro Vector is an advanced vectorization model optimized for high-precision design production and brand asset creation. It generates scalable vectors with nuanced control over line quality, geometry simplification, fills, and color regions. The model is tailored for designers and creative teams seeking production-ready vector outputs for illustration, advertising, UI assets, and print layouts.","capabilities":["io:text-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-pro-vector.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-pro-vector.png","pricingRates":[{"amount":0.3,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.30","amount":0.3,"latencyMs":29337,"exampleId":"deep-sea-prosthetic-socket-wrap"},{"configuration":"vectorize","price":"$0.30","amount":0.3,"latencyMs":19612,"exampleId":"koji-starter-culture-pouch-label"},{"configuration":"vectorize","price":"$0.30","amount":0.3,"latencyMs":33540,"exampleId":"urban-bat-roost-retrofit-explainer"},{"configuration":"vectorize","price":"$0.30","amount":0.3,"latencyMs":30178,"exampleId":"closed-loop-geothermal-system-infographic"}]},{"model":"recraft-v4-vector","air":"recraft:v4@vector","name":"Recraft V4 Vector","creator":"recraft","status":"live","releasedAt":"2026-02-17T00:00:00Z","weight":3,"headline":"Professional vectorization and design-aware image conversion","description":"Recraft V4 Vector is a professional vectorization model that transforms raster images into high-quality scalable vector art. Built for design and marketing workflows, it preserves structural fidelity, contour accuracy, and visual style while producing clean paths and optimized shapes suitable for branding, illustration, and layout production.","capabilities":["io:text-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/recraft-v4-vector.jpg","ogImage":"https://assets.runware.ai/og/recraft-v4-vector.png","pricingRates":[{"amount":0.08,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize · 1024×1024","price":"$0.08","amount":0.08,"latencyMs":18900,"exampleId":"urban-apiary-operations-icon-set"},{"configuration":"vectorize · 1536×768","price":"$0.08","amount":0.08,"latencyMs":19690,"exampleId":"geothermal-utility-campaign-key-visual"},{"configuration":"vectorize · 1536×768","price":"$0.08","amount":0.08,"latencyMs":25411,"exampleId":"cryogenic-specimen-logistics-icon-set"},{"configuration":"vectorize · 1024×1024","price":"$0.08","amount":0.08,"latencyMs":19220,"exampleId":"avalanche-forecaster-persona-portrait"}]},{"model":"tripo-v3-1","air":"tripo:v3.1@0","name":"Tripo 3D v3.1","creator":"tripo","status":"live","releasedAt":"2026-02-11T00:00:00Z","weight":3,"headline":"High-fidelity 3D asset generation with enhanced geometry and texture precision","description":"Tripo 3D v3.1 is a high-fidelity 3D generation model that creates production-ready 3D assets from text prompts or images. It delivers enhanced detail, stronger geometry precision, and improved texture quality compared to earlier versions, with support for PBR materials and export in GLB, FBX, OBJ, and STL formats.","capabilities":["io:text-to-3d","io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/tripo-v3-1.jpg","ogImage":"https://assets.runware.ai/og/tripo-v3-1.png","pricingOverview":"Pricing starts $0.3 per generation, for Text-to-3D, and $0.4 for Image-to-3D and Multiview-to-3D","pricingRates":[{"amount":0.3,"unit":"output","label":"Text-to-3D"},{"amount":0.4,"unit":"output","label":"Image-to-3D"}],"pricingExamples":[{"configuration":"segmented-3d","price":"$0.30","amount":0.3,"latencyMs":111043,"exampleId":"near-future-apiary-smoker-prop"},{"configuration":"reference-to-3d","price":"$0.30","amount":0.3,"latencyMs":238081,"exampleId":"metro-signal-relay-room"},{"configuration":"geometry-only-3d","price":"$0.30","amount":0.3,"latencyMs":106090,"exampleId":"polar-research-ice-core-auger"},{"configuration":"quad-mesh","price":"$0.35","amount":0.35,"latencyMs":123572,"exampleId":"sculptural-watering-can-product-model"},{"configuration":"text-to-3d","price":"$0.50","amount":0.5,"latencyMs":213532,"exampleId":"wildfire-ranger-pangolin-character"}]},{"model":"tencent-hunyuan-3d-3-1-pro","air":"tencent:hunyuan-3d@3.1-pro","name":"Hunyuan 3D 3.1 Pro","status":"live","releasedAt":"2026-02-10T00:00:00Z","weight":3,"headline":"High-fidelity 3D asset generation with multi-view reconstruction and PBR materials","description":"Hunyuan 3D 3.1 Pro is a production-grade 3D generation model from Tencent that creates high-fidelity assets from text prompts or images. It uses enhanced 8-view reconstruction for improved spatial accuracy and reduced geometric artifacts, producing clean topology, watertight meshes, and PBR materials suitable for game development, 3D printing, and professional animation workflows.","capabilities":["io:text-to-3d","io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/tencent-hunyuan-3d-3-1-pro.jpg","ogImage":"https://assets.runware.ai/og/tencent-hunyuan-3d-3-1-pro.png","pricingOverview":"Pricing is per call starting at $0.225 for geometry, $0.375 for Normal. \nThe add-on are cumulative as they stack on top of the base task price when those parameters are used for $0.150 per add on : MultiViewImages, EnablePBR and FaceCount","pricingRates":[{"amount":0.225,"unit":"output","label":"Geometry"},{"amount":0.375,"unit":"output","label":"Normal"},{"amount":0.525,"unit":"output","label":"Normal + EnableBPR"}],"pricingExamples":[{"configuration":"text-to-3d-geometry-only","price":"$0.375","amount":0.375,"latencyMs":81717,"exampleId":"pediatric-radiology-axolotl-mascot"},{"configuration":"image-to-3d-geometry-only","price":"$0.375","amount":0.375,"latencyMs":137248,"exampleId":"polar-telegraph-cabin-environment"},{"configuration":"image-to-3d","price":"$0.525","amount":0.525,"latencyMs":128141,"exampleId":"electric-tufting-gun-ecommerce-model"},{"configuration":"text-to-3d","price":"$0.525","amount":0.525,"latencyMs":135924,"exampleId":"runway-friction-tester-trailer"},{"configuration":"text-to-3d-pbr","price":"$0.675","amount":0.675,"latencyMs":180634,"exampleId":"modern-beekeeping-smoker-product-model"},{"configuration":"image-to-3d-pbr","price":"$0.825","amount":0.825,"latencyMs":214406,"exampleId":"wetland-ranger-muskrat-character-model"}]},{"model":"tencent-hunyuan-3d-3-1-rapid","air":"tencent:hunyuan-3d@3.1-rapid","name":"Hunyuan 3D 3.1 Rapid","status":"live","releasedAt":"2026-02-10T00:00:00Z","weight":3,"headline":"Fast 3D asset generation optimized for rapid prototyping and iteration","description":"Hunyuan 3D 3.1 Rapid is a speed-optimized 3D generation model from Tencent designed for fast prototyping and iteration. It generates 3D assets from text prompts or single images in seconds, producing clean meshes with PBR materials. Ideal for concept evaluation, rapid asset library building, and early-stage design workflows where speed is prioritized over maximum geometric detail.","capabilities":["io:text-to-3d","io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/tencent-hunyuan-3d-3-1-rapid.jpg","ogImage":"https://assets.runware.ai/og/tencent-hunyuan-3d-3-1-rapid.png","pricingOverview":"Pricing is per call starting at $0.225 for both Text-to-3D and Image-to-3D.\nThe add-on EnablePBR is at $0.150 on top of the price if this parameters is used","pricingRates":[{"amount":0.225,"unit":"output","label":"Text-to-3D"},{"amount":0.225,"unit":"output","label":"Image-to-3D"},{"amount":0.375,"unit":"output","label":"Image-to-3D + EnablePBR"}],"pricingExamples":[{"configuration":"image-to-3d-geometry-only","price":"$0.225","amount":0.225,"latencyMs":22416,"exampleId":"pediatric-mri-training-cabin"},{"configuration":"image-to-3d","price":"$0.225","amount":0.225,"latencyMs":62265,"exampleId":"abyssal-salvage-sonar-beacon"},{"configuration":"text-to-3d","price":"$0.225","amount":0.225,"latencyMs":107347,"exampleId":"touring-opera-storm-machine-set"},{"configuration":"text-to-3d-geometry-only","price":"$0.225","amount":0.225,"latencyMs":19512,"exampleId":"soil-museum-mole-mascot"},{"configuration":"image-to-3d-pbr","price":"$0.375","amount":0.375,"latencyMs":67897,"exampleId":"antarctic-seed-vault-entry-set"},{"configuration":"text-to-3d-pbr","price":"$0.375","amount":0.375,"latencyMs":71458,"exampleId":"1970s-pool-admissions-kiosk"}]},{"model":"alibaba-qwen-image-2-0","air":"alibaba:qwen-image@2.0","name":"Qwen-Image-2.0","creator":"alibaba","status":"live","releasedAt":"2026-02-10T00:00:00Z","weight":3,"headline":"Unified image generation and editing with professional text rendering","description":"Qwen-Image-2.0 is an advanced image generation and editing model from Alibaba that produces high-quality images at native 2K resolution and renders professional-grade text within visuals. It unifies text-to-image and image-to-image editing into a single model with strong semantic understanding and adheres to detailed prompt instructions. The model excels at generating images that include complex textual content, infographics, posters, and layout-driven visuals.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-2-0.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-2-0.png","pricingRates":[{"amount":0.035,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 1536×1024","price":"$0.035","amount":0.035,"latencyMs":17161,"exampleId":"museum-textile-freezing-explainer"},{"configuration":"text-to-image · 1024×1024","price":"$0.035","amount":0.035,"latencyMs":3977,"exampleId":"nocturnal-tidepool-survey-poster"},{"configuration":"reference-to-image · 1536×1024","price":"$0.035","amount":0.035,"latencyMs":16649,"exampleId":"cloud-seeding-documentary-thumbnail"},{"configuration":"text-to-image · 1024×1024","price":"$0.035","amount":0.035,"latencyMs":11801,"exampleId":"harbor-crane-operator-portrait"}]},{"model":"vidu-q3-turbo","air":"vidu:4@2","name":"Vidu Q3 Turbo","creator":"vidu","status":"live","releasedAt":"2026-02-09T00:00:00Z","weight":3,"headline":"Low-latency multimodal video generation with native audio","description":"Vidu Q3 Turbo is a speed-optimized multimodal video generation model that produces short video clips with synchronized audio directly from text or images. It prioritizes fast inference and responsive iteration while preserving stable motion, coherent composition, and reliable audio alignment, making it suitable for rapid prototyping and production workflows where latency is critical.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/vidu-q3-turbo.jpg","ogImage":"https://assets.runware.ai/og/vidu-q3-turbo.png","pricingOverview":"Pricing starts at $0.026/s at 540p, $0.039/s at 720p, and $0.052/s at 1080p.  ","pricingRates":[{"amount":0.02275,"unit":"durationSecond","label":"540p"},{"amount":0.026,"unit":"durationSecond","label":"360p"},{"amount":0.03575,"unit":"durationSecond","label":"720p"},{"amount":0.04225,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 960×960 · 5s","price":"$0.179","amount":0.17875,"latencyMs":200752,"exampleId":"pelagic-signal-logo-reveal"},{"configuration":"text-to-video · 720p · 6s · audio","price":"$0.214","amount":0.2145,"latencyMs":48499,"exampleId":"aquaculture-net-cleaner-product-demo"},{"configuration":"text-to-video · 720p · 6s · audio","price":"$0.214","amount":0.2145,"latencyMs":65382,"exampleId":"precision-seed-meter-product-explainer"},{"configuration":"first-frame · 720p · 8s · audio","price":"$0.286","amount":0.286,"latencyMs":67931,"exampleId":"vintage-pen-restoration-reel"}]},{"model":"klingai-image-3-0","air":"klingai:kling-image@3","name":"Kling IMAGE 3.0","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":3,"headline":"2K to 4K image generation with improved realism and practical image-to-image editing","description":"Kling IMAGE 3.0 is an image generation model that targets professional-grade outputs with native 2K to 4K resolution. It focuses on realism through stronger handling of textures, lighting, and materials, and it supports image-to-image workflows for iterative refinement of subjects or layouts while keeping results consistent.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-image-3-0.jpg","ogImage":"https://assets.runware.ai/og/klingai-image-3-0.png","pricingRates":[{"amount":0.028,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1360×768","price":"$0.028","amount":0.028,"latencyMs":29342,"exampleId":"solar-cold-storage-social-ad"},{"configuration":"reference-to-image · 2k","price":"$0.028","amount":0.028,"latencyMs":56620,"exampleId":"container-cargo-sweat-prevention-explainer"},{"configuration":"reference-to-image · 2k","price":"$0.028","amount":0.028,"latencyMs":44831,"exampleId":"seed-library-fundraiser-shirt-graphic"},{"configuration":"text-to-image · 1360×768","price":"$0.028","amount":0.028,"latencyMs":44953,"exampleId":"geothermal-techno-playlist-cover"}]},{"model":"klingai-video-3-0-omni-pro","air":"klingai:kling-video@o3-pro","name":"Kling VIDEO 3.0 Omni Pro","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":3,"headline":"Unified multimodal video generation with native audio and higher-fidelity renders","description":"Kling VIDEO 3.0 Omni Pro is a unified multimodal video model that generates HD clips from text or images with native audio output. It prioritizes detail, motion realism, and stable subject identity, and it supports reference-driven generation plus prompt-based video editing with strong temporal consistency.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-o3-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-o3-pro.png","pricingOverview":"Pricing starts from $0.112/s without audio. ","pricingRates":[{"amount":0.112,"unit":"durationSecond","label":"1080p · no video input + no audio"},{"amount":0.14,"unit":"durationSecond","label":"1080p · no video input + audio"},{"amount":0.168,"unit":"durationSecond","label":"1080p · video input + no audio"}],"pricingExamples":[{"configuration":"first-frame · 5s","price":"$0.70","amount":0.7,"latencyMs":104458,"exampleId":"glassblowing-linocut-heritage-film"},{"configuration":"first-last-frame · 7s","price":"$0.784","amount":0.784,"latencyMs":161634,"exampleId":"jacquard-loom-music-video-interlude"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.84","amount":0.84,"latencyMs":214553,"exampleId":"museum-diver-background-composite"},{"configuration":"elements · 1080p · 6s","price":"$0.84","amount":0.84,"latencyMs":181156,"exampleId":"oyster-farm-salinity-meter-reel"},{"configuration":"video-edit · 1080p","price":"$1.01","amount":1.008,"latencyMs":234473,"exampleId":"racing-wheelchair-violet-colorway-film"},{"configuration":"multi-shot · 1080p · 8s","price":"$1.12","amount":1.12,"latencyMs":182625,"exampleId":"midnight-laundromat-drumline-sequence"},{"configuration":"reference-to-video · 1440p · 8s","price":"$1.34","amount":1.344,"latencyMs":308188,"exampleId":"seaside-lido-seasonal-preview-film"}]},{"model":"klingai-video-3-0-omni-standard","air":"klingai:kling-video@o3-standard","name":"Kling VIDEO 3.0 Omni Standard","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":3,"headline":"Cost-efficient multimodal video generation with native audio and editing","description":"Kling VIDEO 3.0 Omni Standard is a cost-efficient version of the 3.0 Omni generation that produces HD video from text or images with native audio. It balances quality with speed and price, and it supports reference-based generation plus prompt-based video edits that preserve temporal stability across the clip.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-o3-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-o3-standard.png","pricingRates":[{"amount":0.084,"unit":"durationSecond","label":"720p · no input + no audio"},{"amount":0.126,"unit":"durationSecond","label":"720p · video input + no audio"},{"amount":0.112,"unit":"durationSecond","label":"720p · audio + no input"}],"pricingExamples":[{"configuration":"first-frame · 6s","price":"$0.504","amount":0.504,"latencyMs":64956,"exampleId":"urban-bat-census-logo-loop"},{"configuration":"text-to-video · 720p · 5s","price":"$0.56","amount":0.56,"latencyMs":52899,"exampleId":"linocut-steelworks-photo-restyle"},{"configuration":"first-last-frame · 8s","price":"$0.672","amount":0.672,"latencyMs":81291,"exampleId":"adaptive-rain-cape-virtual-try-on"},{"configuration":"video-edit · 720p","price":"$0.756","amount":0.756,"latencyMs":132778,"exampleId":"self-sealing-bicycle-tire-explainer"},{"configuration":"elements · 720p · 10s","price":"$0.84","amount":0.84,"latencyMs":166658,"exampleId":"alpine-tram-seasonal-transition-film"},{"configuration":"reference-to-video · 960×960 · 7s","price":"$0.882","amount":0.882,"latencyMs":161419,"exampleId":"manuscript-scanner-product-demonstration"},{"configuration":"multi-shot · 720p · 8s","price":"$0.896","amount":0.896,"latencyMs":79259,"exampleId":"cranberry-trail-vfx-cleanup"}]},{"model":"klingai-video-3-0-pro","air":"klingai:kling-video@3-pro","name":"Kling VIDEO 3.0 Pro","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":3,"headline":"High-fidelity multimodal video generation with native audio and advanced editing","description":"Kling VIDEO 3.0 Pro is a unified multimodal video model that generates high-quality video with synchronized audio from text or images. It supports reference-guided generation, prompt-based editing, fine control over motion and pacing, and stable temporal coherence for cinematic and narrative clips. Native audio output includes dialogue, ambient sound, and effects aligned to the visuals.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-3-0-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-3-0-pro.png","pricingOverview":"Pricing starts at $0.112/s without audio & $0.168/s with audio. ","pricingRates":[{"amount":0.112,"unit":"durationSecond","label":"1080p · no audio"},{"amount":0.168,"unit":"durationSecond","label":"1080p · audio"},{"amount":0.168,"unit":"durationSecond","label":"1080p · motion control"}],"pricingExamples":[{"configuration":"first-frame · 6s","price":"$0.672","amount":0.672,"latencyMs":99259,"exampleId":"precision-seed-singulator-demo"},{"configuration":"elements · 7s","price":"$0.784","amount":0.784,"latencyMs":436421,"exampleId":"drained-pool-art-pop-music-video"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.84","amount":0.84,"latencyMs":208905,"exampleId":"passive-floodgate-mechanism-demo"},{"configuration":"first-last-frame · 8s","price":"$0.896","amount":0.896,"latencyMs":201794,"exampleId":"telecommunications-documentary-title-sequence"},{"configuration":"reference-to-video","price":"$1.01","amount":1.008,"latencyMs":285831,"exampleId":"rare-book-conservator-briefing"},{"configuration":"multi-shot · 1080p · 9s","price":"$1.51","amount":1.512,"latencyMs":203736,"exampleId":"koji-culture-training-presenter-clip"}]},{"model":"klingai-video-3-0-standard","air":"klingai:kling-video@3-standard","name":"Kling VIDEO 3.0 Standard","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":3,"headline":"Multimodal video generation with native audio and efficient performance","description":"Kling VIDEO 3.0 Standard generates synchronized video and audio from text and images with a balance of quality, speed, and cost. It supports reference-based generation and prompt-driven edits while maintaining temporal stability and clear motion. Native audio output includes dialogue and ambient sound that aligns with the visual content.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-3-0-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-3-0-standard.png","pricingOverview":"Pricing starts at $0.084/s without audio & $0.126/s with audio. ","pricingRates":[{"amount":0.084,"unit":"durationSecond","label":"720p · no audio"},{"amount":0.126,"unit":"durationSecond","label":"720p · audio"},{"amount":0.126,"unit":"durationSecond","label":"720p · motion control"},{"amount":0.084,"unit":"durationSecond","label":"720p"}],"pricingExamples":[{"configuration":"first-frame · 5s","price":"$0.63","amount":0.63,"latencyMs":80177,"exampleId":"rare-seed-archive-motion-ident"},{"configuration":"first-last-frame · 8s","price":"$0.672","amount":0.672,"latencyMs":77861,"exampleId":"sea-salt-harvest-music-video"},{"configuration":"text-to-video · 720p · 6s","price":"$0.756","amount":0.756,"latencyMs":91932,"exampleId":"automatic-flood-barrier-product-demo"},{"configuration":"reference-to-video","price":"$0.756","amount":0.756,"latencyMs":168933,"exampleId":"museum-textile-quarantine-briefing"},{"configuration":"elements · 7s","price":"$0.882","amount":0.882,"latencyMs":458340,"exampleId":"airfield-wildlife-patrol-b-roll"},{"configuration":"multi-shot · 960×960 · 12s","price":"$1.01","amount":1.008,"latencyMs":119794,"exampleId":"geothermal-survey-logo-ident"}]},{"model":"heygen-video-agent","air":"heygen:video-agent@0","name":"HeyGen Video Agent","creator":"heygen","zdr":true,"status":"live","releasedAt":"2026-02-01T00:00:00Z","weight":3,"headline":"AI-powered prompt-to-video production with avatars, B-roll, and motion graphics","description":"HeyGen Video Agent is an AI video production model that generates complete, multi-scene videos from a single text prompt. It automates the full production pipeline — scriptwriting, avatar selection, shot planning, B-roll integration, motion graphics, captions, and editing — producing broadcast-ready videos with consistent branding. The agent supports customizable avatars, voice cloning, and iterative editing without full regeneration, enabling scalable video content creation for marketing, training, and social media.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/heygen-video-agent.jpg","ogImage":"https://assets.runware.ai/og/heygen-video-agent.png","pricingRates":[{"amount":0.097,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"scripted-photo-avatar · 10s","price":"$0.84","amount":0.8403,"latencyMs":554462,"exampleId":"pocket-soil-sensor-explainer"},{"configuration":"text-to-video · 15s","price":"$1.13","amount":1.1291,"latencyMs":410146,"exampleId":"modern-beekeeper-apparel-showcase-reel"},{"configuration":"text-to-video · 15s","price":"$1.47","amount":1.4713,"latencyMs":564062,"exampleId":"river-edna-sampler-product-explainer"},{"configuration":"scripted-stock-avatar · 20s","price":"$1.47","amount":1.4713,"latencyMs":405576,"exampleId":"tidal-foundry-lyric-video-segment"}]},{"model":"sourceful-riverflow-2-0-fast","air":"sourceful:riverflow-2.0@fast","name":"Riverflow 2.0 Fast","creator":"sourceful","status":"live","releasedAt":"2026-01-30T00:00:00Z","weight":3,"headline":"Fast production image generation with reference-based super resolution and font control","description":"Riverflow 2.0 Fast is an optimized image generation and editing model designed for latency-sensitive production pipelines. It maintains strong prompt adherence, accurate product rendering via reference-based super resolution, and dependable font control while prioritizing speed and throughput for large-scale brand and advertising workflows.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sourceful-riverflow-2-0-fast.jpg","ogImage":"https://assets.runware.ai/og/sourceful-riverflow-2-0-fast.png","pricingBasis":"passthrough","pricingRates":[{"amount":0.02,"unit":"output","label":"1K"},{"amount":0.04,"unit":"output","label":"2K"}],"pricingExamples":[{"configuration":"text-to-image · 1248×832","price":"$0.02","amount":0.02,"latencyMs":33629,"exampleId":"fossil-lab-landscape-expansion"},{"configuration":"text-to-image · 1248×832","price":"$0.02","amount":0.02,"latencyMs":39544,"exampleId":"smart-apiary-monitor-lifestyle-shot"},{"configuration":"reference-to-image · 2K","price":"$0.04","amount":0.04,"latencyMs":93977,"exampleId":"brutalist-museum-cafe-menu-cover"},{"configuration":"detail-enhancement · 2K","price":"$0.24","amount":0.24,"latencyMs":299791,"exampleId":"pigeon-racing-documentary-series-poster"}]},{"model":"sourceful-riverflow-2-0-pro","air":"sourceful:riverflow-2.0@pro","name":"Riverflow 2.0 Pro","creator":"sourceful","status":"live","releasedAt":"2026-01-30T00:00:00Z","weight":3,"headline":"Professional image generation with reference-based super resolution and font control","description":"Riverflow 2.0 Pro is a professional image generation and editing model built for high-accuracy commercial workflows. It delivers consistent layouts, precise product rendering through reference-based super resolution, and reliable font control for brand-critical typography. A multi-stage generation and self-correction process reduces visual errors and enables production-ready output for ads, ecommerce, packaging, and editorial content.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sourceful-riverflow-2-0-pro.jpg","ogImage":"https://assets.runware.ai/og/sourceful-riverflow-2-0-pro.png","pricingBasis":"passthrough","pricingRates":[{"amount":0.15,"unit":"output","label":"1K"},{"amount":0.15,"unit":"output","label":"2K"},{"amount":0.33,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"text-to-image · 1248×832","price":"$0.15","amount":0.15,"latencyMs":141059,"exampleId":"dung-beetle-navigation-illustration"},{"configuration":"reference-to-image · 2K","price":"$0.15","amount":0.15,"latencyMs":69654,"exampleId":"municipal-conspiracy-series-poster"},{"configuration":"text-to-image · 1248×832","price":"$0.15","amount":0.15,"latencyMs":133624,"exampleId":"abyssal-courier-game-key-art"},{"configuration":"detail-enhancement · 4K","price":"$0.53","amount":0.53,"latencyMs":565512,"exampleId":"apiary-sensor-lifestyle-photograph"}]},{"model":"vidu-q3","air":"vidu:4@1","name":"Vidu Q3","creator":"vidu","status":"live","releasedAt":"2026-01-30T00:00:00Z","weight":3,"headline":"Multimodal video generation with native audio and intelligent shot planning","description":"Vidu Q3 is a multimodal video generation model that creates video with synchronized audio directly from text or images, supports intelligent multi-shot sequencing, and produces complete outputs with stable visuals and embedded subtitles without post-processing.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/vidu-q3.jpg","ogImage":"https://assets.runware.ai/og/vidu-q3.png","pricingOverview":"From $0.0455 per sec","pricingRates":[{"amount":0.02925,"unit":"durationSecond","label":"540p"},{"amount":0.0325,"unit":"durationSecond","label":"360p"},{"amount":0.065,"unit":"durationSecond","label":"720p"},{"amount":0.078,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 540p · 8s","price":"$0.234","amount":0.234,"latencyMs":121288,"exampleId":"museum-conservation-presenter-update"},{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.325","amount":0.325,"latencyMs":109954,"exampleId":"avalanche-airbag-deployment-explainer"},{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.325","amount":0.325,"latencyMs":154718,"exampleId":"coffee-cherry-sorter-demo"},{"configuration":"first-frame · 720p · 6s","price":"$0.39","amount":0.39,"latencyMs":90522,"exampleId":"mobile-farrier-service-reel"}]},{"model":"inworld-tts-1-5-max","air":"inworld:tts@1.5-max","name":"Inworld TTS-1.5 Max","creator":"inworld","status":"live","releasedAt":"2026-01-21T00:00:00Z","weight":3,"headline":"High-fidelity expressive text-to-speech with rich prosody and multilingual support","description":"Inworld TTS-1.5 Max is a high-fidelity text-to-speech model engineered for expressive voice synthesis with rich prosody, nuanced emotional range, and broadcast-ready audio quality. It supports a wide set of languages and delivers more natural pronunciation and expressive variation suitable for narration, content creation, and immersive character voices. The Max variant prioritizes audio quality and expressiveness while still supporting responsive generation.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/inworld-tts-1-5-max.jpg","ogImage":"https://assets.runware.ai/og/inworld-tts-1-5-max.png","pricingOverview":"$0.05 / 1000 characters","pricingRates":[{"amount":0.00005,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.00455","amount":0.00455,"latencyMs":2834,"exampleId":"championship-pinball-broadcast-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.00515","amount":0.00515,"latencyMs":2954,"exampleId":"seed-library-vocal-audio-logo"},{"configuration":"text-to-audio","price":"$0.0251","amount":0.02515,"latencyMs":7808,"exampleId":"antarctic-field-lab-pa-bed"},{"configuration":"text-to-audio","price":"$0.0297","amount":0.0297,"latencyMs":8716,"exampleId":"urban-forestry-podcast-cold-open"},{"configuration":"text-to-audio","price":"$0.0297","amount":0.0297,"latencyMs":9137,"exampleId":"cold-storage-energy-explainer-voiceover"}]},{"model":"inworld-tts-1-5-mini","air":"inworld:tts@1.5-mini","name":"Inworld TTS-1.5 Mini","creator":"inworld","status":"live","releasedAt":"2026-01-21T00:00:00Z","weight":3,"headline":"Low-latency expressive text-to-speech optimized for real-time apps","description":"Inworld TTS-1.5 Mini is a lightweight text-to-speech model designed for real-time voice experiences with ultra-low latency and efficient performance. It delivers natural, expressive audio suitable for interactive agents, voice assistants, and conversational applications where responsiveness is critical. The Mini variant balances speed and quality, enabling responsive speech output even under constrained compute conditions.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/inworld-tts-1-5-mini.jpg","ogImage":"https://assets.runware.ai/og/inworld-tts-1-5-mini.png","pricingOverview":"$0.025 / 1000 characters","pricingRates":[{"amount":0.000025,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.00267","amount":0.002675,"latencyMs":1757,"exampleId":"dockside-fishmonger-vocal-ambience"},{"configuration":"text-to-audio","price":"$0.00363","amount":0.003625,"latencyMs":2145,"exampleId":"forensic-accounting-podcast-stinger"},{"configuration":"text-to-audio","price":"$0.00645","amount":0.00645,"latencyMs":3456,"exampleId":"cold-war-numbers-station-bed"},{"configuration":"text-to-audio","price":"$0.0121","amount":0.012125,"latencyMs":4149,"exampleId":"film-archive-preservation-lesson"}]},{"model":"meshy-6","air":"meshy:meshy@6","name":"Meshy-6","creator":"meshy","status":"live","releasedAt":"2026-01-18T00:00:00Z","weight":3,"headline":"High-fidelity 3D generation with cleaner geometry, low-poly workflows, and production-ready mesh control","description":"Meshy-6 is a 3D generation model for text-to-3D and image-to-3D workflows. It is built for cleaner geometry, sharper hard-surface detail, low-poly generation, configurable quad or triangle topology, polycount and symmetry control, A/T pose support, and textured outputs with optional PBR materials. It fits game asset creation, 3D printing, and production-oriented 3D pipelines.","capabilities":["io:text-to-3d","io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/meshy-6.jpg","ogImage":"https://assets.runware.ai/og/meshy-6.png","pricingOverview":"Pricing is $0.8 per generation, for Text-to-3D, Image-to-3D, and Multiple Image To 3D","pricingRates":[{"amount":0.8,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-3d","price":"$0.80","amount":0.8,"latencyMs":359789,"exampleId":"arctic-seed-vault-airlock-diorama"},{"configuration":"multi-image-to-3d","price":"$0.80","amount":0.8,"latencyMs":231910,"exampleId":"botanical-specimen-press-product-model"},{"configuration":"image-to-3d","price":"$0.80","amount":0.8,"latencyMs":378532,"exampleId":"planetarium-control-booth-diorama"},{"configuration":"reference-to-3d","price":"$0.80","amount":0.8,"latencyMs":241359,"exampleId":"pediatric-hydrotherapy-pool-room"}],"guides":[{"slug":"generating-3d","title":"Generating 3D models from prompts and photos","description":"How to generate 3D models with Meshy-6 from a prompt or photos: choosing text vs image input, framing a clean reference, using multiple views, and prompting for 3D form.","date":"2026-06-16","url":"https://runware.ai/docs/models/meshy-6/guides/generating-3d"}]},{"model":"bria-fibo-edit-tools","air":"bria:21@2","name":"Bria Fibo Edit Tools","creator":"bria","status":"live","releasedAt":"2026-01-16T00:00:00Z","weight":3,"headline":"Unified image editing foundation for colorize, relight, restore, blend, reseason, sketch-to-image, and expand","description":"Bria Fibo Edit Tools is an image editing model that groups seven dedicated tools, each applied to a single source image. The tools are colorize (recolor a photo in contemporary, vivid, black-and-white, or sepia tones), relight (change the light type and direction), restore (repair old or degraded photos), blend (merge a prompted element, material, or style into the image), reseason (move the scene to spring, summer, autumn, or winter), sketch-to-image (turn a sketch into a finished image), and expand (extend the canvas beyond the original borders). Blend and expand are guided by a text prompt, while the other tools work from the image alone or with a preset choice.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-fibo-edit-tools.jpg","ogImage":"https://assets.runware.ai/og/bria-fibo-edit-tools.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"relight","price":"$0.04","amount":0.04,"latencyMs":21819,"exampleId":"chiaroscuro-artifact-catalog-portrait"},{"configuration":"blend","price":"$0.04","amount":0.04,"latencyMs":24361,"exampleId":"floating-wetland-keynote-backdrop"},{"configuration":"restore","price":"$0.04","amount":0.04,"latencyMs":21839,"exampleId":"saturation-diver-archive-headshot"},{"configuration":"sketch-to-image","price":"$0.04","amount":0.04,"latencyMs":17335,"exampleId":"panoramic-seafloor-trencher-concept-art"},{"configuration":"reseason","price":"$0.04","amount":0.04,"latencyMs":24206,"exampleId":"prefab-sauna-winter-variant"},{"configuration":"expand","price":"$0.04","amount":0.04,"latencyMs":10224,"exampleId":"smart-beehive-panorama-extension"},{"configuration":"colorize","price":"$0.04","amount":0.04,"latencyMs":52542,"exampleId":"neon-workshop-archive-colorization"}]},{"model":"bfl-flux-2-klein-4b","air":"runware:400@4","name":"FLUX.2 [klein] 4B","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-15T00:00:00Z","weight":3,"headline":"Fastest Klein model for real-time image generation and editing","description":"FLUX.2 [klein] 4B is a 4-step distilled image generation and editing model optimized for ultra-low latency inference. It delivers near real-time performance with strong visual quality, enabling interactive workflows and responsive production systems on more constrained hardware.","architecture":"flux-2-klein-4b","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-klein-4b.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-klein-4b.png","pricingMeasured":[{"configuration":"1024x1024","price":0.0006}],"pricingExamples":[{"configuration":"text-to-image · 720p","price":"$0.0006","amount":0.0006,"latencyMs":1301,"exampleId":"community-seed-swap-shirt-graphic"},{"configuration":"text-to-image · 720p","price":"$0.0006","amount":0.0006,"latencyMs":1514,"exampleId":"nordic-crime-series-key-art"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0019","amount":0.0019,"latencyMs":5879,"exampleId":"center-pivot-linocut-wall-print"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0019","amount":0.0019,"latencyMs":6485,"exampleId":"pigeon-racing-documentary-poster"}]},{"model":"bfl-flux-2-klein-4b-base","air":"runware:400@5","name":"FLUX.2 [klein] 4B Base","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-15T00:00:00Z","weight":3,"headline":"Compact undistilled model for efficient image generation and editing","description":"FLUX.2 [klein] 4B Base is a compact undistilled image generation and editing model with an exceptional quality-to-size ratio. It is well suited for local deployment, efficient fine-tuning, and custom pipelines that require flexibility on limited hardware.","architecture":"flux-2-klein-base-4b","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-klein-4b-base.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-klein-4b-base.png","pricingMeasured":[{"configuration":"1024x1024","price":0.0019}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.0013","amount":0.0013,"latencyMs":3244,"exampleId":"roman-mosaic-survey-clean-plate"},{"configuration":"text-to-image · 1024×1024","price":"$0.0013","amount":0.0013,"latencyMs":3355,"exampleId":"lodz-industrial-tourism-social-poster"},{"configuration":"reference-to-image · 768×1024","price":"$0.0032","amount":0.0032,"latencyMs":9499,"exampleId":"sleeper-train-breakfast-menu"},{"configuration":"reference-to-image · 768×1024","price":"$0.0032","amount":0.0032,"latencyMs":10100,"exampleId":"cold-chain-workwear-campaign-visual"}]},{"model":"bfl-flux-2-klein-9b-base","air":"runware:400@3","name":"FLUX.2 [klein] 9B Base","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-15T00:00:00Z","weight":3,"headline":"Undistilled foundation model for high-quality image generation and editing","description":"FLUX.2 [klein] 9B Base is the undistilled foundation model of the Klein family, offering full model capacity for image generation and editing. It is optimized for fine-tuning, customization, and post-training workflows where flexibility, control, and maximum training signal are required.","architecture":"flux-2-klein-base-9b","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-klein-9b-base.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-klein-9b-base.png","pricingMeasured":[{"configuration":"1024x1024","price":0.00416}],"pricingExamples":[{"configuration":"text-to-image · 1344×768","price":"$0.00338","amount":0.00338,"latencyMs":3994,"exampleId":"harbor-pilot-picture-book-spread"},{"configuration":"text-to-image · 1344×768","price":"$0.00338","amount":0.00338,"latencyMs":4065,"exampleId":"avalanche-rescue-team-portrait-composite"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0091","amount":0.0091,"latencyMs":16222,"exampleId":"coastal-harvest-workwear-lookbook"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0091","amount":0.0091,"latencyMs":17244,"exampleId":"fog-harvester-campaign-key-visual"}]},{"model":"alibaba-wan2-6-flash","air":"alibaba:wan@2.6-flash","name":"Wan2.6 Flash","creator":"alibaba","status":"live","releasedAt":"2026-01-15T00:00:00Z","weight":3,"headline":"Fast distilled image-to-video generation model","description":"Wan2.6 Flash is a distilled, low-latency variant of the Wan2.6 multimodal video model designed for rapid image to video generation with fluid motion, visual stability, and optional synchronized audio. It produces HD clips from detailed static images while preserving subject structure and motion realism, making it suitable for preview workflows and high-throughput creative pipelines.","capabilities":["io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-6-flash.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-6-flash.png","pricingOverview":"A 1-second clip starts at $0.025 (no audio) & $0.05 (audio) at 720p. To use a video as a reference, it starts at $0.025/s (no audio, 720p) applied on top of base pricing.","pricingRates":[{"amount":0.05,"unit":"durationSecond","label":"720p · audio"},{"amount":0.025,"unit":"durationSecond","label":"720p · no audio"},{"amount":0.075,"unit":"durationSecond","label":"1080p · audio"},{"amount":0.0375,"unit":"durationSecond","label":"1080p · no audio"},{"amount":0.05,"unit":"durationSecond","label":"720p · R2V · audio"},{"amount":0.025,"unit":"durationSecond","label":"720p · R2V · no audio"},{"amount":0.075,"unit":"durationSecond","label":"1080p · R2V · audio"},{"amount":0.0375,"unit":"durationSecond","label":"1080p · R2V · no audio"}],"pricingExamples":[{"configuration":"first-frame · 1080p · 5s · audio","price":"$0.376","amount":0.3755,"latencyMs":50003,"exampleId":"battery-sorting-safety-presenter-clip"},{"configuration":"first-frame · 1080p · 5s · audio","price":"$0.376","amount":0.3755,"latencyMs":49869,"exampleId":"subsea-network-documentary-title"},{"configuration":"reference-to-video · 720p · 8s · audio","price":"$0.40","amount":0.4,"latencyMs":96929,"exampleId":"cryogenic-wildlife-biobank-b-roll"},{"configuration":"audio-driven · 720p · 10s","price":"$0.501","amount":0.5011,"latencyMs":50425,"exampleId":"nitrate-film-safety-presenter"}]},{"model":"lightricks-ltx-2","air":"lightricks:ltx@2","name":"LTX-2","creator":"lightricks","zdr":true,"status":"live","releasedAt":"2026-01-05T00:00:00Z","weight":3,"headline":"Open-source AI video model with synchronized audio and high-fidelity output","description":"LTX-2 is an open-source multimodal video foundation model that generates synchronized video and audio from text or image prompts. It produces high-quality motion sequences with native 4K resolution and smooth temporal coherence, making it suitable for creative video generation, production workflows, and audiovisual storytelling.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/lightricks-ltx-2.jpg","ogImage":"https://assets.runware.ai/og/lightricks-ltx-2.png","license":"https://ltx.video/api-license-agreement","pricingRates":[{"amount":0.03,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-video · 1080p · 5s","price":"$0.15","amount":0.15,"latencyMs":262923,"exampleId":"heritage-seed-documentary-title-sting"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.15","amount":0.15,"latencyMs":262902,"exampleId":"adaptive-running-blade-social-reel"},{"configuration":"first-frame · 720p · 8s","price":"$0.24","amount":0.24,"latencyMs":210429,"exampleId":"technical-beekeeper-suit-showcase-reel"},{"configuration":"first-frame · 720p · 8s","price":"$0.24","amount":0.24,"latencyMs":190790,"exampleId":"sculptural-copper-handbag-showcase"}]},{"model":"ace-step-v1-5-base","air":"runware:ace-step@v1.5-base","name":"ACE-Step v1.5 Base","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-01T00:00:00Z","weight":3,"headline":"Open-source music generation with voice cloning, lyric editing, and multilingual support","description":"ACE-Step v1.5 Base is an open-source music generation foundation model built on a hybrid LLM planner and Diffusion Transformer architecture. It generates full tracks from text prompts with support for voice cloning, lyric editing, remixing, cover generation, and compositions up to 10 minutes. It supports over 50 languages and runs on consumer hardware with under 4GB VRAM.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ace-step-v1-5-base.jpg","ogImage":"https://assets.runware.ai/og/ace-step-v1-5-base.png","pricingOverview":"$0.00015 per second of audio","pricingMeasured":[{"configuration":"duration 60 · steps 40","price":0.00195,"latencyMs":4772},{"configuration":"duration 30 · steps 100","price":0.00285,"latencyMs":5371},{"configuration":"duration 60 · steps 100","price":0.0039,"latencyMs":6651},{"configuration":"duration 60 · steps 300","price":0.0096,"latencyMs":13129}],"pricingExamples":[{"configuration":"text-to-audio · 60s","price":"$0.0039","amount":0.0039,"latencyMs":13474,"exampleId":"ceramic-heat-battery-explainer-song"},{"configuration":"audio-extend","price":"$0.0087","amount":0.0087,"latencyMs":22305,"exampleId":"coastal-lighting-retrofit-explainer-narration"},{"configuration":"audio-repaint","price":"$0.0115","amount":0.01155,"latencyMs":29847,"exampleId":"oyster-farm-insurance-explainer-voiceover"},{"configuration":"audio-driven","price":"$0.0163","amount":0.01635,"latencyMs":31667,"exampleId":"avalanche-beacon-training-narration"}]},{"model":"ace-step-v1-5-turbo","air":"runware:ace-step@v1.5-turbo","name":"ACE-Step v1.5 Turbo","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-01T00:00:00Z","weight":3,"headline":"Fast music generation optimized for speed with reduced inference steps","description":"ACE-Step v1.5 Turbo is a speed-optimized variant of the ACE-Step v1.5 music generation model. It delivers faster inference with fewer denoising steps while retaining the core capabilities of the Base model, including voice cloning, lyric editing, remixing, and multilingual support across 50+ languages.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ace-step-v1-5-turbo.jpg","ogImage":"https://assets.runware.ai/og/ace-step-v1-5-turbo.png","pricingOverview":"$0.0001 per second of audio","pricingMeasured":[{"configuration":"duration 60 · steps 10","price":0.0009,"latencyMs":3370},{"configuration":"duration 300 · steps 10","price":0.00195,"latencyMs":10563}],"pricingExamples":[{"configuration":"audio-repaint","price":"$0.0009","amount":0.0009,"latencyMs":12699,"exampleId":"museum-hazard-training-narration"},{"configuration":"text-to-audio · 60s","price":"$0.0009","amount":0.0009,"latencyMs":8432,"exampleId":"ocean-observatory-podcast-theme-music"},{"configuration":"text-to-audio · 60s","price":"$0.0009","amount":0.0009,"latencyMs":5955,"exampleId":"closed-loop-dyehouse-explainer-voiceover"},{"configuration":"audio-driven","price":"$0.0039","amount":0.0039,"latencyMs":16998,"exampleId":"arctic-seed-vault-soundscape"}]},{"model":"alibaba-qwen3-tts-1-7b-base","air":"alibaba:qwen@3-tts-1.7b-base","name":"Qwen3-TTS 1.7B Base","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-01T00:00:00Z","weight":3,"headline":"High-quality multilingual text-to-speech with voice cloning and ultra-low latency","description":"Qwen3-TTS 1.7B Base is the foundation text-to-speech model from Alibaba's Qwen3-TTS family. It generates human-like speech across 10+ languages including Chinese, English, Japanese, Korean, and European languages. It supports voice cloning from a 3-second audio sample and achieves latency as low as 97ms for real-time applications.","capabilities":["io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen3-tts-1-7b-base.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen3-tts-1-7b-base.png","pricingRates":[{"amount":0.000015,"unit":"character"}],"pricingExamples":[{"configuration":"speaker-embedding","price":"$0.0005","amount":0.0005,"latencyMs":3772,"exampleId":"orbital-greenhouse-intercom-ambience"},{"configuration":"speaker-embedding","price":"$0.0028","amount":0.0028,"latencyMs":4814,"exampleId":"coffee-grounds-collection-explainer"},{"configuration":"voice-cloning","price":"$0.0029","amount":0.0029,"latencyMs":4286,"exampleId":"rare-seed-licensing-podcast-intro"},{"configuration":"voice-cloning","price":"$0.0052","amount":0.0052,"latencyMs":3608,"exampleId":"urban-leak-detection-explainer"}]},{"model":"alibaba-qwen3-tts-1-7b-voicedesign","air":"alibaba:qwen@3-tts-1.7b-voicedesign","name":"Qwen3-TTS 1.7B VoiceDesign","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-01T00:00:00Z","weight":3,"headline":"Text-to-speech with voice creation from natural language descriptions","description":"Qwen3-TTS 1.7B VoiceDesign is a text-to-speech model from Alibaba that creates custom voices from natural language descriptions specifying emotion, tone, and prosody. It supports voice cloning from a 3-second audio sample, generates speech in 10+ languages including Chinese, English, Japanese, Korean, and European languages, and achieves latency as low as 97ms.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen3-tts-1-7b-voicedesign.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen3-tts-1-7b-voicedesign.png","pricingRates":[{"amount":0.000015,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.0042","amount":0.0042,"latencyMs":3975,"exampleId":"semiconductor-cleanroom-intercom-ambience"},{"configuration":"text-to-audio","price":"$0.0063","amount":0.0063,"latencyMs":5580,"exampleId":"municipal-water-leak-explainer-voiceover"},{"configuration":"text-to-audio","price":"$0.0078","amount":0.0078,"latencyMs":6392,"exampleId":"shellfish-farm-monitoring-explainer"},{"configuration":"text-to-audio","price":"$0.0108","amount":0.0108,"latencyMs":8681,"exampleId":"forensic-entomology-training-narration"}]},{"model":"twinflow-z-image-turbo","air":"runware:twinflow-z-image-turbo@0","name":"TwinFlow Z-Image-Turbo","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-12-29T00:00:00Z","weight":3,"headline":"Fast text to image generation with turbo inference","description":"TwinFlow Z-Image-Turbo is an image generation model optimized for fast inference. It supports text-to-image synthesis producing high-quality results with low latency for rapid iteration workflows.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/twinflow-z-image-turbo.jpg","ogImage":"https://assets.runware.ai/og/twinflow-z-image-turbo.png","pricingMeasured":[{"configuration":"1024x1024 · 4 steps","price":0.0006}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0006","amount":0.0006,"latencyMs":1305,"exampleId":"canine-blood-donor-social-ad"},{"configuration":"text-to-image","price":"$0.0006","amount":0.0006,"latencyMs":1803,"exampleId":"hatchery-biosecurity-training-illustration"},{"configuration":"text-to-image","price":"$0.0006","amount":0.0006,"latencyMs":1865,"exampleId":"forensic-entomologist-profile-avatar"},{"configuration":"text-to-image","price":"$0.0006","amount":0.0006,"latencyMs":2378,"exampleId":"antarctic-logistics-strategy-game-art"}]},{"model":"alibaba-qwen-image-layered","air":"alibaba:qwen-image@layered","name":"Qwen-Image-Layered","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-12-19T00:00:00Z","weight":3,"headline":"Layered image decomposition model for editable RGBA outputs","description":"Qwen-Image-Layered decomposes a static image into multiple RGBA layers, enabling independent editing of semantically distinct components without interfering with other parts of the image. This layered representation supports high-fidelity image editing tasks like resizing, repositioning, recoloring, and object manipulation with consistent detail and transparency handling.","architecture":"qwen-image-layered","capabilities":["io:image-to-image","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-layered.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-layered.png","pricingMeasured":[{"configuration":"28 steps","price":0.0275}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":26151,"exampleId":"historic-observatory-photo-cleanup"},{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":25850,"exampleId":"fossil-skull-background-separation"},{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":27350,"exampleId":"heritage-railway-crew-portrait"},{"configuration":"reference-to-image","price":"$0.0154","amount":0.0154,"latencyMs":29679,"exampleId":"urban-beekeeping-screenprint-illustration"}]},{"model":"microsoft-trellis-2","air":"microsoft:trellis-2@4b","name":"TRELLIS.2","creator":"microsoft","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-12-17T00:00:00Z","weight":3,"headline":"High-fidelity image-to-3D generative model with compact structured latents","description":"TRELLIS.2 is a 4-billion-parameter 3D generative model that transforms 2D images into fully textured 3D assets with complex topology, sharp features, and physically based rendering materials. It uses a novel field-free sparse voxel representation and scalable structured latent space to enable efficient, high-resolution 3D generation from single image input.","capabilities":["io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/microsoft-trellis-2.jpg","ogImage":"https://assets.runware.ai/og/microsoft-trellis-2.png","pricingOverview":"Pricing is based on processing time. Most generations cost just a few cents, depending on complexity.","pricingMeasured":[{"configuration":"settings.resolution 512","price":0.0077,"latencyMs":14253},{"configuration":"settings.texSlat.steps 40","price":0.0179,"latencyMs":37869},{"configuration":"settings.sparseStructure.steps 40","price":0.0186,"latencyMs":32779},{"configuration":"settings.textureSize 4096","price":0.0243,"latencyMs":41784},{"configuration":"settings.shapeSlat.steps 40","price":0.0256,"latencyMs":43871}],"pricingExamples":[{"configuration":"image-to-3d","price":"$0.009","amount":0.009,"latencyMs":21322,"exampleId":"portable-farrier-hoof-stand-model"},{"configuration":"image-to-3d","price":"$0.0166","amount":0.0166,"latencyMs":30485,"exampleId":"underground-safety-mole-character-model"},{"configuration":"image-to-3d","price":"$0.0218","amount":0.0218,"latencyMs":78785,"exampleId":"neon-laundromat-cutaway-set-module"},{"configuration":"image-to-3d","price":"$0.0269","amount":0.0269,"latencyMs":45628,"exampleId":"tidal-monitoring-cabin-environment-module"}]},{"model":"alibaba-wan2-6","air":"alibaba:wan@2.6","name":"Wan2.6","creator":"alibaba","status":"live","releasedAt":"2025-12-16T00:00:00Z","weight":3,"headline":"Multimodal video generation with multi-shot and native sound","description":"Wan2.6 is a multimodal video model for text to video and image to video generation with support for multi-shot sequencing and native sound. It emphasizes temporal stability, consistent visual structure across shots, and reliable alignment between visuals and audio in short form video generation.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-6.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-6.png","pricingOverview":"$0.10 per second at 720p and $0.15 per second at 1080p.","pricingRates":[{"amount":0.1,"unit":"durationSecond","label":"720p"},{"amount":0.15,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"reference-to-video · 1088×832 · 5s","price":"$0.50","amount":0.5,"latencyMs":98348,"exampleId":"heritage-seed-documentary-title-sequence"},{"configuration":"text-to-video · 1080p · 5s · audio","price":"$0.751","amount":0.7511,"latencyMs":111082,"exampleId":"inline-magnetic-separator-demo"},{"configuration":"audio-driven · 960×960 · 10s","price":"$1.00","amount":1.0022,"latencyMs":111240,"exampleId":"avalanche-patrol-series-teaser"},{"configuration":"first-frame · 1080p · 10s","price":"$1.50","amount":1.5033,"latencyMs":95605,"exampleId":"marine-robotics-logo-loop"},{"configuration":"multi-shot · 1080p · 15s · audio","price":"$2.26","amount":2.2557,"latencyMs":172285,"exampleId":"archive-flood-recovery-presenter-clip"}]},{"model":"sync-react-1","air":"sync:react-1@1","name":"react-1","creator":"sync","status":"live","releasedAt":"2025-12-12T00:00:00Z","weight":3,"headline":"Video performance editing from audio or written direction","description":"react-1 is a video performance editing model designed for post-production direction without reshoots. It modifies acting and emotional delivery within existing footage while preserving identity and visual continuity, enabling directors to reshape performances using audio or written guidance.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sync-react-1.jpg","ogImage":"https://assets.runware.ai/og/sync-react-1.png","pricingOverview":"$0.1467 per second","pricingRates":[{"amount":0.1467,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven","price":"$0.941","amount":0.94132505,"latencyMs":287598,"exampleId":"arctic-seed-vault-mystery-teaser"},{"configuration":"audio-driven","price":"$0.966","amount":0.96577495,"latencyMs":401622,"exampleId":"night-radio-mascot-logo-loop"},{"configuration":"audio-driven","price":"$0.99","amount":0.990225,"latencyMs":34756,"exampleId":"apiary-scanner-product-demo"},{"configuration":"audio-driven","price":"$1.11","amount":1.11247495,"latencyMs":153424,"exampleId":"municipal-leak-detector-demo"},{"configuration":"audio-driven","price":"$1.39","amount":1.39365,"latencyMs":165340,"exampleId":"cacao-free-chocolate-reaction-reel"},{"configuration":"audio-driven","price":"$1.44","amount":1.43643755,"latencyMs":255340,"exampleId":"infrared-art-scanner-demo"}]},{"model":"klingai-video-2-6-standard","air":"klingai:kling-video@2.6-standard","name":"Kling VIDEO 2.6 Standard","creator":"klingai","status":"live","releasedAt":"2025-12-11T00:00:00Z","weight":3,"headline":"High-quality AI video generation with strong motion and camera control","description":"Kling VIDEO 2.6 Standard is a high-quality AI video generation model focused on producing visually coherent short clips with stable motion, expressive camera movement, and strong prompt adherence. It generates video from text prompts or an optional input image, making it suitable for cinematic previews, social content, and creative prototyping where audio is not required.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-video-2-6-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-2-6-standard.png","pricingRates":[{"amount":0.14,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"reference-to-video","price":"$0.33","amount":0.33,"latencyMs":195057,"exampleId":"museum-pest-training-presenter-clip"},{"configuration":"reference-to-video","price":"$0.33","amount":0.33,"latencyMs":187315,"exampleId":"apiary-biosecurity-presenter-bulletin"},{"configuration":"text-to-video","price":"$0.70","amount":0.7,"latencyMs":46452,"exampleId":"cryogenic-seed-vault-b-roll"},{"configuration":"text-to-video","price":"$0.70","amount":0.7,"latencyMs":52347,"exampleId":"wine-fraud-docuseries-title-teaser"}]},{"model":"klingai-avatar-2-0-pro","air":"klingai:avatar@2.0-pro","name":"KlingAI Avatar 2.0 Pro","creator":"klingai","status":"live","releasedAt":"2025-12-04T00:00:00Z","weight":3,"headline":"High fidelity avatar video generation with smoother motion and quality","description":"KlingAI Avatar 2.0 Pro builds on the Standard version with higher visual fidelity, smoother motion, and improved expressivity. It generates up to five-minute avatar videos from a single image and audio track, with enhanced detail and production-ready results for varied character types.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-avatar-2-0-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-avatar-2-0-pro.png","pricingOverview":"$0.087 per second","pricingRates":[{"amount":0.087,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven-photo-avatar","price":"$0.348","amount":0.348,"latencyMs":191822,"exampleId":"municipal-compost-talking-logo-ident"},{"configuration":"audio-driven-photo-avatar","price":"$0.441","amount":0.440742,"latencyMs":180222,"exampleId":"avalanche-forecast-stream-motion-ident"},{"configuration":"audio-driven-photo-avatar","price":"$0.696","amount":0.696,"latencyMs":261868,"exampleId":"seaweed-fiber-rain-shell-showcase-reel"},{"configuration":"audio-driven-photo-avatar","price":"$0.719","amount":0.719142,"latencyMs":284680,"exampleId":"coral-nursery-technician-stock-clip"}]},{"model":"klingai-avatar-2-0-standard","air":"klingai:avatar@2.0-standard","name":"KlingAI Avatar 2.0 Standard","creator":"klingai","status":"live","releasedAt":"2025-12-04T00:00:00Z","weight":3,"headline":"Expressive avatar video generation from image and audio","description":"KlingAI Avatar 2.0 Standard generates talking avatar videos from a single portrait image and audio, preserving identity and producing natural lip-sync and expressive motion. It supports up to five minutes of video with multilingual control and gesture clarity for human or cartoon characters.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-avatar-2-0-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-avatar-2-0-standard.png","pricingOverview":"$0.044 per second","pricingRates":[{"amount":0.044,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven-photo-avatar","price":"$0.199","amount":0.199452,"latencyMs":143228,"exampleId":"seed-vault-living-emblem"},{"configuration":"audio-driven-photo-avatar","price":"$0.422","amount":0.4224,"latencyMs":274555,"exampleId":"veterinary-poison-safety-reel"},{"configuration":"audio-driven-photo-avatar","price":"$0.434","amount":0.434104,"latencyMs":233017,"exampleId":"volcano-docuseries-teaser-sequence"},{"configuration":"audio-driven-photo-avatar","price":"$0.868","amount":0.868252,"latencyMs":313665,"exampleId":"shellfish-hatchery-sensor-explainer"}]},{"model":"klingai-image-o1","air":"klingai:kling-image@o1","name":"Kling IMAGE O1","creator":"klingai","status":"live","releasedAt":"2025-12-02T00:00:00Z","weight":3,"headline":"High control AI image model for consistent characters","description":"Kling IMAGE O1 is a high control image generation model for stable characters and precise edits. It supports detailed composition control, strong style handling, and localized modifications without structural drift. Ideal for pipelines that need repeatable shots and complex visual continuity.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-image-o1.jpg","ogImage":"https://assets.runware.ai/og/klingai-image-o1.png","pricingRates":[{"amount":0.028,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image · 1248×832","price":"$0.028","amount":0.028,"latencyMs":45001,"exampleId":"varroa-mite-test-kit-carton"},{"configuration":"reference-to-image · 2k","price":"$0.028","amount":0.028,"latencyMs":31399,"exampleId":"mobile-library-archive-photo-restoration"},{"configuration":"reference-to-image · 2k","price":"$0.028","amount":0.028,"latencyMs":48174,"exampleId":"lowrider-bicycle-builder-editorial-portrait"},{"configuration":"text-to-image · 1248×832","price":"$0.028","amount":0.028,"latencyMs":29833,"exampleId":"estuary-electronic-music-festival-poster"}]},{"model":"kandinsky-5-0-image-lite","air":"runware:kandinsky-5.0-image-lite@1","name":"Kandinsky 5.0 Image Lite","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-26T00:00:00Z","weight":3,"headline":"Efficient text-to-image and image-to-image editing model","description":"Kandinsky 5.0 Image Lite is an open-source image model that generates high-quality images from text prompts and performs image-to-image edits using a reference image. It blends strong prompt adherence with visual fidelity across diverse visual concepts.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/kandinsky-5-0-image-lite.jpg","ogImage":"https://assets.runware.ai/og/kandinsky-5-0-image-lite.png","pricingMeasured":[{"configuration":"1024×1024 · steps 10","price":0.0026,"latencyMs":4725},{"configuration":"1024×1024 · steps 20","price":0.0051,"latencyMs":8602},{"configuration":"1024×1024 · steps 50","price":0.0122,"latencyMs":19114}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0051","amount":0.0051,"latencyMs":35004,"exampleId":"synthesizer-technician-profile-avatar"},{"configuration":"text-to-image","price":"$0.0051","amount":0.0051,"latencyMs":20109,"exampleId":"pneumatic-waste-system-explainer"},{"configuration":"text-to-image","price":"$0.0051","amount":0.0051,"latencyMs":8802,"exampleId":"anime-oyster-hatchery-recruitment-illustration"},{"configuration":"text-to-image","price":"$0.0051","amount":0.0051,"latencyMs":27284,"exampleId":"winter-lido-redevelopment-marketing-visual"}]},{"model":"bfl-flux-2-flex","air":"bfl:6@1","name":"FLUX.2 [flex]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2025-11-25T00:00:00Z","weight":3,"headline":"Configurable FLUX.2 Flex for precise text aligned images","description":"FLUX.2 [flex] is a configurable text to image and image editing model built for precise text placement and stable layouts. It exposes sampling and guidance controls and supports up to ten reference images for consistent characters or products across complex compositions.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-2-flex.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-2-flex.png","pricingOverview":"~$0.1431 per megapixel","pricingRates":[{"amount":0.06,"unit":"outputMegapixel"},{"amount":0.06,"unit":"inputMegapixel"}],"pricingExamples":[{"configuration":"text-to-image · 1536×1024","price":"$0.12","amount":0.12,"latencyMs":36208,"exampleId":"mine-water-heating-explainer"},{"configuration":"text-to-image · 1536×1024","price":"$0.12","amount":0.12,"latencyMs":21984,"exampleId":"subsea-cable-repair-feature-illustration"},{"configuration":"reference-to-image · 1024×1536","price":"$0.24","amount":0.24,"latencyMs":21656,"exampleId":"polar-logistics-report-cover"},{"configuration":"reference-to-image · 1024×1536","price":"$0.24","amount":0.24,"latencyMs":21347,"exampleId":"typefoundry-scratchboard-editorial-portrait"}]},{"model":"dia2-2b","air":"runware:dia2@2b","name":"Dia2 2B","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-19T00:00:00Z","weight":3,"headline":"Streaming dialogue TTS with voice cloning, non-verbal cues, and multi-speaker support","description":"Dia2 2B is a 2 billion parameter streaming text-to-speech model from Nari Labs designed for real-time conversational AI. It begins generating audio immediately from partial text input, supports multi-speaker dialogue via speaker tags, voice cloning from a few seconds of reference audio, and non-verbal cues like laughter, sighs, and coughs. Released under Apache 2.0 for commercial use.","capabilities":["io:text-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/dia2-2b.jpg","ogImage":"https://assets.runware.ai/og/dia2-2b.png","pricingMeasured":[{"configuration":"10 words","price":0.0057,"latencyMs":9137},{"configuration":"100 words","price":0.0183,"latencyMs":23677}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.0048","amount":0.0048,"latencyMs":13720,"exampleId":"alpine-apiary-podcast-stinger"},{"configuration":"text-to-audio","price":"$0.0163","amount":0.01635,"latencyMs":26118,"exampleId":"rare-film-archive-podcast-opening"},{"configuration":"multi-speaker-dialogue","price":"$0.0211","amount":0.02115,"latencyMs":31344,"exampleId":"electric-ferry-vocal-brand-stinger"},{"configuration":"voice-cloning","price":"$0.0355","amount":0.03555,"latencyMs":51867,"exampleId":"mushroom-farm-biosecurity-module-narration"}]},{"model":"bytedance-video-upscaler","air":"bytedance:50@1","name":"ByteDance Video Upscaler","creator":"bytedance","status":"live","releasedAt":"2025-11-01T00:00:00Z","weight":3,"headline":"Enterprise video upscaling for clean 4K restoration","description":"ByteDance Video Upscaler boosts video resolution to 1080p, 2K, or 4K with advanced denoising and motion enhancement. It restores color, reduces compression artifacts, and improves clarity for legacy films, UGC clips, and short narrative content through a simple API.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-video-upscaler.jpg","ogImage":"https://assets.runware.ai/og/bytedance-video-upscaler.png","pricingOverview":"Pricing starts at $0.00265 per second for 720p under 30fps. Resolution and Fps highly affects pricing ","pricingRates":[{"amount":0.00265,"unit":"durationSecond","label":"720p · <30 fps"},{"amount":0.0053,"unit":"durationSecond","label":"1080p · <30 fps"},{"amount":0.0053,"unit":"durationSecond","label":"720p · >30fps"},{"amount":0.01061,"unit":"durationSecond","label":"1080p · >30fps"}],"pricingExamples":[{"configuration":"upscale","price":"$0.129","amount":0.1285,"latencyMs":133674,"exampleId":"deep-sea-expedition-footage-restoration"},{"configuration":"upscale","price":"$0.129","amount":0.1285,"latencyMs":130674,"exampleId":"seed-vault-animated-logo-ident"},{"configuration":"upscale","price":"$0.129","amount":0.1285,"latencyMs":110063,"exampleId":"antique-watch-restoration-reel"},{"configuration":"upscale","price":"$0.129","amount":0.1285,"latencyMs":118768,"exampleId":"algae-dyed-rain-shell-showcase"}]},{"model":"klingai-2-5-turbo-standard","air":"klingai:6@0","name":"KlingAI 2.5 Turbo Standard","creator":"klingai","status":"live","releasedAt":"2025-10-23T00:00:00Z","weight":3,"headline":"Fast cinematic image to video generation for creators","description":"KlingAI 2.5 Turbo Standard is a streamlined image to video model tuned for speed and cost efficiency. It generates smooth cinematic clips with strong motion control and clear frames at up to 720p. Ideal for rapid iteration in creative pipelines and production tests.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-2-5-turbo-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-2-5-turbo-standard.png","pricingOverview":"Each generation will cost $0.0420/s for 720p.","pricingRates":[{"amount":0.042,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-video · 5s","price":"$0.21","amount":0.21,"latencyMs":70644,"exampleId":"boutique-dental-crown-reel"},{"configuration":"text-to-video · 5s","price":"$0.21","amount":0.21,"latencyMs":53944,"exampleId":"silent-film-archive-documentary-teaser"},{"configuration":"first-frame · 10s","price":"$0.42","amount":0.42,"latencyMs":89487,"exampleId":"textile-dye-archivist-presenter-clip"},{"configuration":"first-frame · 10s","price":"$0.42","amount":0.42,"latencyMs":86427,"exampleId":"urban-beekeeping-outerwear-showcase-reel"}]},{"model":"vidu-q2-turbo","air":"vidu:3@2","name":"Vidu Q2 Turbo","creator":"vidu","status":"live","releasedAt":"2025-10-21T00:00:00Z","weight":3,"headline":"Faster Vidu Q2 video generation with advanced motion control","description":"Vidu Q2 Turbo is the fast tier of the Q2 video model. It targets rapid iteration for creative pipelines. It keeps the cinematic look of Vidu Q2 Pro. It adds shorter latency, stronger large motion control, and smoother camera movement for prompt driven video shots.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-q2-turbo.jpg","ogImage":"https://assets.runware.ai/og/vidu-q2-turbo.png","pricingRates":[{"amount":0.066,"unit":"output","label":"480p · 8s"},{"amount":0.0825,"unit":"output","label":"720p · 4s"},{"amount":0.1375,"unit":"output","label":"720p · 8s"},{"amount":0.2475,"unit":"output","label":"1080p · 8s"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s","price":"$0.11","amount":0.11,"latencyMs":59330,"exampleId":"pollinator-documentary-vertical-title-teaser"},{"configuration":"reference-to-video · 960×960 · 6s","price":"$0.138","amount":0.1375,"latencyMs":50777,"exampleId":"autonomous-beach-cleaner-b-roll"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.193","amount":0.1925,"latencyMs":59924,"exampleId":"antarctic-meteorite-curator-briefing"}]},{"model":"anthropic-claude-haiku-4-5","air":"anthropic:claude@haiku-4.5","name":"Claude Haiku 4.5","creator":"anthropic","endpoints":["chat-completions","messages"],"status":"live","releasedAt":"2025-10-15T00:00:00Z","weight":3,"headline":"Fast, cost-efficient multimodal language model for low-latency agents and scaled reasoning workloads","description":"Claude Haiku 4.5 is Anthropic's fastest and most cost-efficient Claude model. It is built for latency-sensitive applications, high-volume agents, sub-agent orchestration, coding assistance, and budget-conscious deployments that still need strong reasoning and multimodal understanding.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/anthropic-claude-haiku-4-5.jpg","ogImage":"https://assets.runware.ai/og/anthropic-claude-haiku-4-5.png","contextWindow":200000,"maxOutputTokens":64000,"pricingBasis":"token","pricingRates":[{"amount":0.000001,"unit":"inputToken"},{"amount":0.000005,"unit":"outputToken"},{"amount":1e-7,"unit":"cachedInputToken"},{"amount":0.00000125,"unit":"cacheWriteToken"}],"pricingExamples":[{"configuration":"text-generation","price":"$0.00473","amount":0.004725,"latencyMs":9203,"exampleId":"wrong-item-refund-resolution"},{"configuration":"text-generation","price":"$0.0051","amount":0.005102,"latencyMs":10839,"exampleId":"saas-incident-executive-brief"},{"configuration":"text-generation","price":"$0.00658","amount":0.00658,"latencyMs":10271,"exampleId":"restaurant-refrigeration-bid-decision-memo"},{"configuration":"reference-to-text","price":"$0.00991","amount":0.009909,"latencyMs":27900,"exampleId":"hotel-room-defect-punch-list"}]},{"model":"bytedance-seedream-4-0","air":"bytedance:5@0","name":"Seedream 4.0","creator":"bytedance","status":"live","releasedAt":"2025-09-09T00:00:00Z","weight":3,"headline":"High speed 4K AI image generation and editing model","description":"Seedream 4.0 is ByteDance’s multimodal image model for fast 2K to 4K generation. It supports text prompts, image editing with natural language, and multi image reference. It maintains style consistency across batches and handles bilingual Chinese and English workflows.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedream-4-0.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedream-4-0.png","pricingOverview":"Each image generation costs $0.03 at 1024x1024.","pricingRates":[{"amount":0.03,"unit":"output"}],"pricingExamples":[{"configuration":"sequential-images · 2304×1728","price":"$0.03","amount":0.03,"latencyMs":140053,"exampleId":"peatland-firefighting-game-concept-sequence"},{"configuration":"sequential-images · 2304×1728","price":"$0.03","amount":0.03,"latencyMs":110152,"exampleId":"subsea-cable-repair-thumbnail-series"},{"configuration":"text-to-image · 3024×1296","price":"$0.03","amount":0.03,"latencyMs":15967,"exampleId":"aerospace-acoustic-engineer-executive-portrait"},{"configuration":"reference-to-image · 1728×2304","price":"$0.03","amount":0.03,"latencyMs":38538,"exampleId":"coastal-restoration-textile-wall-print"}]},{"model":"bytedance-omnihuman-1-5","air":"bytedance:5@2","name":"OmniHuman-1.5","creator":"bytedance","status":"live","releasedAt":"2025-08-27T00:00:00Z","weight":3,"headline":"Cognitive avatar video from image, audio, and text","description":"OmniHuman-1.5 generates high fidelity avatar video from a single image with audio and optional text prompts. It fuses multimodal reasoning with diffusion motion to keep identity stable, lip sync accurate, and gestures context aware for long, multi subject clips.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-omnihuman-1-5.jpg","ogImage":"https://assets.runware.ai/og/bytedance-omnihuman-1-5.png","pricingRates":[{"amount":0.13247,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"masked-audio-driven-photo-avatar","price":"$1.08","amount":1.08,"latencyMs":235458,"exampleId":"pedal-compost-pickup-social-reel"},{"configuration":"masked-audio-driven-photo-avatar","price":"$1.35","amount":1.3488,"latencyMs":221722,"exampleId":"antarctic-post-office-recruitment-reel"},{"configuration":"audio-driven-photo-avatar","price":"$1.44","amount":1.4448,"latencyMs":356256,"exampleId":"lithium-battery-safety-presenter-clip"},{"configuration":"audio-driven-photo-avatar","price":"$1.83","amount":1.8288,"latencyMs":322860,"exampleId":"livestock-auctioneer-documentary-b-roll"}]},{"model":"bytedance-seedance-1-0-pro","air":"bytedance:2@1","name":"Seedance 1.0 Pro","creator":"bytedance","status":"live","releasedAt":"2025-06-01T00:00:00Z","weight":3,"headline":"Seedance 1.0 Pro high fidelity 1080p text and image to video","description":"Seedance 1.0 Pro is a ByteDance video model for 5 to 10 second clips at up to 1080p. It supports text prompts and image first frames. It delivers smooth motion with strong temporal consistency. Ideal for multi shot storytelling, ads, and design previews in real time pipelines.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seedance-1-0-pro.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seedance-1-0-pro.png","pricingOverview":"Each generation will cost $0.0968/s for 1080p.","pricingRates":[{"amount":0.01925,"unit":"durationSecond","label":"480p"},{"amount":0.09694,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 480p · 8s","price":"$0.155","amount":0.1547667,"latencyMs":66486,"exampleId":"dry-powder-inhaler-mechanism-demo"},{"configuration":"first-frame · 1080p · 5s","price":"$0.489","amount":0.4887432,"latencyMs":49504,"exampleId":"polar-mail-museum-presenter-clip"},{"configuration":"first-frame · 1080p · 5s","price":"$0.489","amount":0.4887432,"latencyMs":68722,"exampleId":"hatchery-biosecurity-presenter-clip"},{"configuration":"text-to-video · 2176×928 · 6s","price":"$0.566","amount":0.5661612,"latencyMs":67016,"exampleId":"mycelium-acoustic-tile-social-reel"}]},{"model":"openai-gpt-image-1-mini","air":"openai:1@2","name":"GPT Image 1 Mini","creator":"openai","status":"deprecated","replacedBy":"openai-gpt-image-2","deactivatesAt":"2026-12-01T00:00:00Z","releasedAt":"2025-04-23T00:00:00Z","weight":3,"headline":"Cost-effective GPT Image 1 variant optimized for speed and efficiency","description":"GPT Image 1 Mini is a lighter variant of OpenAI's GPT Image 1 model. It offers faster generation at a lower cost while retaining core capabilities including text-to-image generation, image editing, and text rendering. It is suited for high-volume workflows, rapid prototyping, and cost-sensitive applications where the full GPT Image 1 model may be excessive.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-image-1-mini.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-image-1-mini.png","pricingRates":[{"amount":0.036,"unit":"output"}],"pricingExamples":[{"configuration":"transparent-background · 1024×1024","price":"$0.036","amount":0.036,"latencyMs":41268,"exampleId":"battery-safety-book-character-sticker"},{"configuration":"inpaint · 1024×1024","price":"$0.036","amount":0.036,"latencyMs":33860,"exampleId":"observatory-architecture-square-outpaint"},{"configuration":"reference-to-image · 1024×1536","price":"$0.052","amount":0.052,"latencyMs":49685,"exampleId":"pipe-organ-workshop-cover"},{"configuration":"text-to-image · 1536×1024","price":"$0.052","amount":0.052,"latencyMs":47158,"exampleId":"polar-research-vessel-christening-invitation"}]},{"model":"hidream-i1-fast","air":"runware:97@3","name":"HiDream-I1 Fast","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-04-07T00:00:00Z","weight":3,"headline":"HiDream-I1 Fast for low latency text to image generation","description":"HiDream-I1 Fast is a distilled text to image model tuned for very low latency workflows. It runs with fewer diffusion steps than Full or Dev variants and keeps strong prompt adherence. Ideal for real-time previews, rapid drafts and bulk image generation in production pipelines.","architecture":"hidream-fast","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/hidream-i1-fast.jpg","ogImage":"https://assets.runware.ai/og/hidream-i1-fast.png","pricingOverview":"Each image generation costs $0.0032 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 16 steps","price":0.0038},{"configuration":"1024x1024 · 22 steps","price":0.0051}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":3199,"exampleId":"opera-archive-heist-poster"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":4084,"exampleId":"seed-vault-atlas-cover"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":3339,"exampleId":"roller-rink-organist-editorial-portrait"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":3779,"exampleId":"cold-chain-logistics-campaign-visual"}]},{"model":"rundiffusion-juggernaut-lightning-flux","air":"rundiffusion:110@101","name":"Juggernaut Lightning Flux","creator":"rundiffusion","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-03-06T00:00:00Z","weight":3,"headline":"Ultra fast Flux-based model for high volume image generation","description":"Juggernaut Lightning Flux is a Flux-based image generator tuned for speed. It delivers high quality outputs with fewer steps. Ideal for mood boards, rapid iteration, and bulk asset creation. Suits solo workflows and production pipelines that need low latency.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/rundiffusion-juggernaut-lightning-flux.jpg","ogImage":"https://assets.runware.ai/og/rundiffusion-juggernaut-lightning-flux.png","pricingOverview":"~$0.0014 per megapixel","pricingMeasured":[{"configuration":"1024x1024","price":0.0017},{"configuration":"512x512","price":0.0008},{"configuration":"768x768","price":0.0008},{"configuration":"1280x1280","price":0.0017},{"configuration":"1536x1536","price":0.0025}],"pricingExamples":[{"configuration":"text-to-image · 1344×768 · steps 4","price":"$0.00078","amount":0.00078,"latencyMs":1408,"exampleId":"setouchi-island-cycling-campaign-key-art"},{"configuration":"expand · 720p","price":"$0.00078","amount":0.00078,"latencyMs":17716,"exampleId":"polar-archivist-profile-avatar"},{"configuration":"image-to-image · 1024×1024 · steps 4","price":"$0.00078","amount":0.00078,"latencyMs":2587,"exampleId":"industrial-mycology-album-cover"},{"configuration":"inpaint · 896×1152 · steps 4","price":"$0.00169","amount":0.00169,"latencyMs":4331,"exampleId":"experimental-civic-noise-album-cover"}]},{"model":"bfl-flux-1-fill-dev","air":"runware:102@1","name":"FLUX.1 Fill [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-11-21T00:00:00Z","weight":3,"headline":"FLUX.1 Fill dev for precise AI image inpainting","description":"FLUX.1 Fill [dev] is an open image editing model for inpainting and outpainting with text guidance and masks. It can replace objects, extend scenes, and adjust regions while preserving context. Ideal for pipelines that need controllable edits on real or generated images.","architecture":"flux-1-dev","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-fill-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-fill-dev.png","pricingMeasured":[{"configuration":"512×512 · steps 20","price":0.0013,"latencyMs":4531},{"configuration":"1024×1024 · steps 20","price":0.0026,"latencyMs":7048},{"configuration":"1024×1024 · steps 40","price":0.0045,"latencyMs":10516},{"configuration":"2048×2048 · steps 20","price":0.0109,"latencyMs":20819}],"pricingExamples":[{"configuration":"expand · 720p","price":"$0.0019","amount":0.0019,"latencyMs":26543,"exampleId":"vintage-snowmobile-auction-landscape"},{"configuration":"inpaint · 1024×1024","price":"$0.0026","amount":0.0026,"latencyMs":13968,"exampleId":"terrazzo-sample-catalog-cleanup"},{"configuration":"inpaint · 1024×1024","price":"$0.0026","amount":0.0026,"latencyMs":11415,"exampleId":"apiary-smoker-enamel-colorway"},{"configuration":"expand · 1344×768 · steps 26","price":"$0.0032","amount":0.0032,"latencyMs":8598,"exampleId":"cathedral-restoration-editorial-banner"}]},{"model":"flux-1-dev-controlnet-union-pro-pose","air":"runware:29@1","name":"FLUX.1 [dev] ControlNet Union Pro – Pose","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":3,"headline":"FLUX.1 ControlNet mode for pose-locked character and human figure generation","description":"FLUX.1 ControlNet Union Pro – Pose is the pose-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is intended for character and figure generation where body pose, limb placement, and overall human structure should follow a pose reference while the prompt determines appearance and style.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-pose.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-pose.png"},{"model":"vit-age-classifier","air":"runware:153@1","name":"ViT Age Classifier","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-07-28T00:00:00Z","weight":3,"headline":"Vision transformer model for estimating age from facial images","description":"ViT Age Classifier is an image classification model that predicts the age category of a person based on a facial image. It uses a Vision Transformer architecture fine-tuned for age estimation to analyze facial features and output structured age predictions.","capabilities":["io:image-to-text","op:age-classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vit-age-classifier.jpg","ogImage":"https://assets.runware.ai/og/vit-age-classifier.png","pricingMeasured":[{"configuration":"Age classification","price":0.0006,"latencyMs":2279}],"pricingExamples":[{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3966,"exampleId":"museum-visitor-age-segment-audit"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":4012,"exampleId":"documentary-casting-age-cohort-check"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":2501,"exampleId":"oral-history-age-cohort-tag"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3175,"exampleId":"stock-portrait-age-metadata-tag"}]},{"model":"birefnet-general","air":"runware:112@5","name":"BiRefNet General","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":3,"headline":"General-purpose object segmentation model in the BiRefNet family","description":"BiRefNet General is trained for broad segmentation use cases, providing stable and consistent object masks across a wide variety of image domains without task-specific specialization.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-general.jpg","ogImage":"https://assets.runware.ai/og/birefnet-general.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3975}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5347,"exampleId":"hand-spliced-marine-fender-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6479,"exampleId":"beekeeper-orchard-background-swap-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":7048,"exampleId":"circus-portrait-restyle-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":4923,"exampleId":"geothermal-engineer-press-headshot"}]},{"model":"stabilityai-stable-diffusion-xl-refiner-1-0","air":"runware:101055@128080","name":"Stable Diffusion XL Refiner 1.0","creator":"stabilityai","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-07-28T00:00:00Z","weight":3,"headline":"Specialized SDXL checkpoint for the final denoising and image refinement stage","description":"Stable Diffusion XL Refiner 1.0 is the refinement checkpoint in the SDXL 1.0 pipeline. It is trained for the final denoising steps on existing SDXL latents or image-to-image inputs, making it useful when a base SDXL generation needs cleaner fine detail, stronger local texture, and a more polished final pass rather than a full standalone generation workflow.","architecture":"sdxl","capabilities":["op:refine","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/stabilityai-stable-diffusion-xl-refiner-1-0.jpg","ogImage":"https://assets.runware.ai/og/stabilityai-stable-diffusion-xl-refiner-1-0.png"},{"model":"stabilityai-stable-diffusion-xl-v1-0-vae-fix","air":"civitai:101055@128078","name":"Stable Diffusion XL v1.0 VAE Fix","creator":"stabilityai","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-07-28T00:00:00Z","weight":3,"headline":"Native 1024px SDXL generation with a corrected VAE for more reliable text-to-image and image-to-image workflows","description":"Stable Diffusion XL v1.0 VAE Fix is the SDXL 1.0 base checkpoint packaged with a corrected VAE for more stable inference. It keeps SDXL's strong prompt understanding, broad visual range, native 1024x1024 generation, and support for both text-to-image and image-to-image workflows, while reducing the artifact and decoding issues associated with the original embedded VAE.","architecture":"sdxl","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/stabilityai-stable-diffusion-xl-v1-0-vae-fix.jpg","ogImage":"https://assets.runware.ai/og/stabilityai-stable-diffusion-xl-v1-0-vae-fix.png","pricingMeasured":[{"configuration":"1024×1024 · steps 20","price":0.0013,"latencyMs":2871},{"configuration":"1024×1024 · steps 50","price":0.0019,"latencyMs":4027},{"configuration":"2048×2048 · steps 20","price":0.0038,"latencyMs":6703}],"pricingExamples":[{"configuration":"text-to-image · 1216×832 · steps 28","price":"$0.0013","amount":0.0013,"latencyMs":3829,"exampleId":"marine-water-tester-lifestyle-photograph"},{"configuration":"image-to-image · 832×1216 · steps 30","price":"$0.0013","amount":0.0013,"latencyMs":3992,"exampleId":"urban-wildlife-officer-persona-portrait"},{"configuration":"expand · 1152×768 · steps 30","price":"$0.0013","amount":0.0013,"latencyMs":4653,"exampleId":"rare-book-conservator-headshot"},{"configuration":"inpaint · 1024×1024 · steps 32","price":"$0.0019","amount":0.0019,"latencyMs":5629,"exampleId":"polar-biobank-badge-logo"}]},{"model":"yolov8n-face","air":"runware:35@1","name":"YOLOv8n Face","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-01-10T00:00:00Z","weight":3,"headline":"Lightweight face mask generation for realistic images","description":"YOLOv8n Face is a lightweight image masking model that detects realistic 2D faces and returns black-and-white masks for each detection. It is designed for fast automated mask generation in inpainting and detail enhancement workflows where throughput matters more than maximum facial precision.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/yolov8n-face.jpg","ogImage":"https://assets.runware.ai/og/yolov8n-face.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3404}],"pricingExamples":[{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3426,"exampleId":"opera-character-makeup-try-on-masks"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3983,"exampleId":"maritime-exhibition-scene-swap-masks"},{"configuration":"face-mask","price":"$0.0013","amount":0.0013,"latencyMs":4978,"exampleId":"independent-perfumers-group-headshot"},{"configuration":"face-mask","price":"$0.0013","amount":0.0013,"latencyMs":5646,"exampleId":"badlands-expedition-background-swap"}]},{"model":"rundiffusion-juggernaut-z-fast","air":"rundiffusion:300@100","name":"Juggernaut Z Fast","creator":"rundiffusion","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-06-11T00:00:00Z","weight":2,"headline":"Fast iteration model for prompt testing, composition exploration, and early visual direction","description":"Juggernaut Z Fast is a RunDiffusion image model built for rapid prototyping in Juggernaut Z workflows. It is a LoRA-distilled variant designed to move quickly through prompt testing, layout exploration, lighting checks, and early composition work before switching to a higher-fidelity final pass.","architecture":"z-image","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/rundiffusion-juggernaut-z-fast.jpg","ogImage":"https://assets.runware.ai/og/rundiffusion-juggernaut-z-fast.png","pricingMeasured":[{"configuration":"512×512 · steps 8","price":0.00078,"latencyMs":1758},{"configuration":"1024×1024 · steps 8","price":0.00247,"latencyMs":3422},{"configuration":"2048×2048 · steps 8","price":0.01079,"latencyMs":13649}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024","price":"$0.00169","amount":0.00169,"latencyMs":2899,"exampleId":"maritime-museum-scene-composite"},{"configuration":"image-to-image · 1280×768","price":"$0.00169","amount":0.00169,"latencyMs":3188,"exampleId":"constructed-wetland-process-infographic"},{"configuration":"inpaint · 768×1024","price":"$0.00247","amount":0.00247,"latencyMs":13995,"exampleId":"airport-sensory-booth-visualization"},{"configuration":"expand · 1536×1024","price":"$0.00338","amount":0.00338,"latencyMs":7026,"exampleId":"agrivoltaic-energy-campaign-key-visual"}]},{"model":"baidu-ernie-image","air":"baidu:ernie-image@0","name":"ERNIE-Image","creator":"baidu","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-15T00:00:00Z","weight":2,"headline":"General-purpose 8B text-to-image model with strong instruction following, text rendering, and structured layout control","description":"ERNIE-Image is Baidu's 8B text-to-image model built on a single-stream Diffusion Transformer architecture. It is designed for strong prompt adherence, reliable text rendering, and structured visual generation, making it well suited to posters, comics, storyboards, multi-panel layouts, and other workflows where content accuracy and composition matter as much as aesthetics. The standard model emphasizes stronger general-purpose capability and instruction fidelity, typically running at around 50 inference steps.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/baidu-ernie-image.jpg","ogImage":"https://assets.runware.ai/og/baidu-ernie-image.png","pricingOverview":"Pricing is highly dependent on number of parameters. \nThe most significant cost multiplier is the use of promptEnhance(settings.promptEnhance) with an average price increase of 178%. Cost also scales approximately linearly with step count, baseline in use cases is 30 steps. ","pricingMeasured":[{"configuration":"512×512 · steps 30","price":0.0038,"latencyMs":19713},{"configuration":"1024×1024 · steps 15","price":0.0051,"latencyMs":24378},{"configuration":"1024×1024 · steps 30","price":0.0096,"latencyMs":27792},{"configuration":"1024×1024 · steps 50","price":0.0154,"latencyMs":43096}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0205","amount":0.0205,"latencyMs":48323,"exampleId":"geothermal-architecture-journal-cover"},{"configuration":"text-to-image","price":"$0.023","amount":0.023,"latencyMs":70781,"exampleId":"moonlit-oyster-farm-dinner-invitation"},{"configuration":"text-to-image","price":"$0.023","amount":0.023,"latencyMs":58621,"exampleId":"falconry-telemetry-kit-packshot"},{"configuration":"text-to-image","price":"$0.0307","amount":0.0307,"latencyMs":79204,"exampleId":"tidal-choral-festival-poster"}]},{"model":"baidu-ernie-image-turbo","air":"baidu:ernie-image@turbo","name":"ERNIE-Image-Turbo","creator":"baidu","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-04-15T00:00:00Z","weight":2,"headline":"Distilled 8-step ERNIE image variant for faster generation with strong aesthetics and efficient prompting","description":"ERNIE-Image-Turbo is Baidu's distilled fast variant of ERNIE-Image. It is optimized for substantially faster generation, typically requiring only 8 inference steps, while retaining relatively comparable performance to the full model in many scenarios. It is best suited to high-throughput ideation, rapid visual iteration, and workflows that prioritize speed and polished aesthetics over the stronger general-purpose instruction fidelity of the base model.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/baidu-ernie-image-turbo.jpg","ogImage":"https://assets.runware.ai/og/baidu-ernie-image-turbo.png","pricingOverview":"Pricing is highly dependent on number of parameters. \nThe most significant cost multiplier is the use of promptEnhance(settings.promptEnhance) with an average price increase of 178%. Cost also scales approximately linearly with step count, baseline in use cases is 8 steps. ","pricingMeasured":[{"configuration":"512×512 · steps 8","price":0.0013,"latencyMs":16577},{"configuration":"1024×1024 · steps 8","price":0.0019,"latencyMs":27877},{"configuration":"2048×2048 · steps 8","price":0.0064,"latencyMs":30388}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":19299,"exampleId":"antique-violin-restoration-blog-hero"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":26743,"exampleId":"volcanologist-social-profile-avatar"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":27350,"exampleId":"regenerative-wool-insulation-key-visual"},{"configuration":"text-to-image","price":"$0.0019","amount":0.0019,"latencyMs":29816,"exampleId":"geothermal-drilling-campaign-key-visual"}]},{"model":"google-veo-3-1-lite","air":"google:veo@3.1-lite","name":"Veo 3.1 Lite","creator":"google","zdr":true,"status":"live","releasedAt":"2026-03-31T00:00:00Z","weight":2,"headline":"Cost-effective video generation at less than half the price of Veo 3.1 Fast","description":"Veo 3.1 Lite is the most cost-effective model in the Veo 3.1 family, designed for high-volume applications requiring rapid iteration. It supports text-to-video and image-to-video generation at 720p or 1080p in landscape and portrait formats, with customizable duration of 4, 6, or 8 seconds. It maintains the same generation speed as Veo 3.1 Fast at less than 50% of the cost, and includes native synchronized audio generation.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/google-veo-3-1-lite.jpg","ogImage":"https://assets.runware.ai/og/google-veo-3-1-lite.png","pricingRates":[{"amount":0.03,"unit":"durationSecond","label":"720p · without audio"},{"amount":0.05,"unit":"durationSecond","label":"720p · with audio"},{"amount":0.05,"unit":"durationSecond","label":"1080p · without audio"},{"amount":0.08,"unit":"durationSecond","label":"1080p · with audio"}],"pricingExamples":[{"configuration":"first-last-frame · 720p · 4s","price":"$0.20","amount":0.2,"latencyMs":47088,"exampleId":"embroidered-folk-lyric-video"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.48","amount":0.48,"latencyMs":75460,"exampleId":"adaptive-cutlery-social-video-ad"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.48","amount":0.48,"latencyMs":75832,"exampleId":"rare-manuscript-digitization-b-roll-clip"},{"configuration":"first-frame · 1080p · 8s","price":"$0.64","amount":0.64,"latencyMs":106646,"exampleId":"couture-hearing-jewelry-reel"}]},{"model":"prunaai-p-video","air":"prunaai:p-video@0","name":"P-Video","creator":"prunaai","status":"live","releasedAt":"2026-02-26T00:00:00Z","weight":2,"headline":"Real-time AI video generation with draft mode and native audio","description":"Pruna P-Video is a real-time AI video generation model designed for fast creative iteration and production workflows. It supports text-to-video, image-to-video, and audio-to-video through a unified endpoint, delivering up to 1080p at 48 FPS with integrated dialogue generation and audio import. The model emphasizes speed, cost efficiency, sequencing consistency across clips, and stable subject identity, making it well suited for brand content, multi-format distribution, and rapid draft-to-refine pipelines.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-video.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-video.png","pricingRates":[{"amount":0.02,"unit":"durationSecond","label":"720p · Standard"},{"amount":0.005,"unit":"durationSecond","label":"720p · Draft"},{"amount":0.04,"unit":"durationSecond","label":"1080p · Standard"},{"amount":0.01,"unit":"durationSecond","label":"1080p · Draft"}],"pricingExamples":[{"configuration":"draft-preview · 720p · 4s","price":"$0.0202","amount":0.0202,"latencyMs":23845,"exampleId":"seaweed-hatchery-seeding-b-roll"},{"configuration":"audio-driven · 720p","price":"$0.181","amount":0.1808,"latencyMs":23117,"exampleId":"canine-blood-donor-welcome-clip"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.202","amount":0.2017,"latencyMs":27163,"exampleId":"tidal-energy-logo-ident"},{"configuration":"audio-driven-image-animation · 720p","price":"$0.221","amount":0.2208,"latencyMs":31339,"exampleId":"cargo-biosecurity-inspection-briefing"},{"configuration":"first-frame · 1080p · 6s","price":"$0.241","amount":0.2408,"latencyMs":52896,"exampleId":"textile-conservator-training-presenter-clip"}]},{"model":"klingai-image-o3","air":"klingai:kling-image@o3","name":"Kling IMAGE O3","creator":"klingai","status":"live","releasedAt":"2026-02-05T00:00:00Z","weight":2,"headline":"4K Omni image generation with strong consistency and reference control","description":"Kling IMAGE O3 is an Omni image model built for high-fidelity text-to-image and image-to-image generation at up to 4K resolution. It supports multi-image reference prompting, series image generation for coherent variations, and optional face-focused element control to keep identity stable across outputs.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-image-o3.jpg","ogImage":"https://assets.runware.ai/og/klingai-image-o3.png","pricingRates":[{"amount":0.028,"unit":"output","label":"1K & 2K"},{"amount":0.056,"unit":"output","label":"4K"}],"pricingExamples":[{"configuration":"reference-to-image · 1664×2496","price":"$0.028","amount":0.028,"latencyMs":253299,"exampleId":"zero-proof-listening-bar-menu-photograph"},{"configuration":"text-to-image · 2496×1664","price":"$0.028","amount":0.028,"latencyMs":31171,"exampleId":"rainwater-rebate-social-campaign-image"},{"configuration":"text-to-image · 2496×1664","price":"$0.028","amount":0.028,"latencyMs":30083,"exampleId":"seaweed-packaging-linkedin-social-ad"},{"configuration":"reference-to-image · 1664×2496","price":"$0.028","amount":0.028,"latencyMs":291630,"exampleId":"iberian-conservas-menu-cover"}]},{"model":"pixverse-v5-6","air":"pixverse:1@7","name":"PixVerse V5.6","creator":"pixverse","status":"live","releasedAt":"2026-01-26T00:00:00Z","weight":2,"headline":"Enhanced cinematic video generation with improved lip-sync and audio realism","description":"PixVerse V5.6 is an upgraded video generation model that improves visual stability, motion clarity, and audio-visual alignment over previous versions. It supports text-to-video and image-to-video generation with optional native audio, delivering more accurate multi-character lip-sync, cleaner motion in complex scenes, and more natural speech and environmental sound for single-shot cinematic outputs.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/pixverse-v5-6.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v5-6.png","pricingRates":[{"amount":0.04714,"unit":"durationSecond","label":"360p · audio"},{"amount":0.02062,"unit":"durationSecond","label":"360p · no audio"},{"amount":0.04714,"unit":"durationSecond","label":"540p · audio"},{"amount":0.02062,"unit":"durationSecond","label":"540p · no audio"},{"amount":0.05304,"unit":"durationSecond","label":"720p · audio"},{"amount":0.02652,"unit":"durationSecond","label":"720p · no audio"},{"amount":0.07072,"unit":"durationSecond","label":"1080p · audio"},{"amount":0.0442,"unit":"durationSecond","label":"1080p · no audio"}],"pricingExamples":[{"configuration":"first-last-frame · 720p · 5s · audio","price":"$0.265","amount":0.2652,"latencyMs":57371,"exampleId":"maritime-satellite-terminal-deployment-demo"},{"configuration":"text-to-video · 720p · 10s · audio","price":"$0.424","amount":0.4243,"latencyMs":100684,"exampleId":"antarctic-ice-core-science-update"},{"configuration":"text-to-video · 720p · 10s · audio","price":"$0.424","amount":0.4243,"latencyMs":72797,"exampleId":"bell-foundry-folk-music-segment"},{"configuration":"first-frame · 1080p · 8s","price":"$0.442","amount":0.442,"latencyMs":199999,"exampleId":"passive-flood-gate-product-demonstration"}]},{"model":"object-eraser","air":"runware:300@1","name":"Object Eraser","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-19T00:00:00Z","weight":2,"headline":"Targeted object removal with mask-based inpainting","description":"Object Eraser is an image editing model that removes selected objects from an image using a provided mask. It performs targeted inpainting to reconstruct missing regions while preserving surrounding structure, lighting, and scene continuity. Optional prompt conditioning allows control over the fill result, enabling texture matching, contextual completion, and alignment with the original image content.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/object-eraser.jpg","ogImage":"https://assets.runware.ai/og/object-eraser.png","pricingMeasured":[{"configuration":"steps 5 · 1024×1024 input","price":0.0019,"latencyMs":22869}],"pricingExamples":[{"configuration":"inpaint","price":"$0.0032","amount":0.0032,"latencyMs":48561,"exampleId":"wildland-commander-headshot-clean-plate"},{"configuration":"inpaint","price":"$0.0038","amount":0.0038,"latencyMs":63421,"exampleId":"opera-scenic-mural-restoration"},{"configuration":"inpaint","price":"$0.0051","amount":0.0051,"latencyMs":62490,"exampleId":"risograph-atrium-photo-restyle"},{"configuration":"inpaint · steps 5","price":"$0.0064","amount":0.0064,"latencyMs":39399,"exampleId":"geothermal-exhibit-panoramic-clean-plate"}]},{"model":"alibaba-qwen-image-2512","air":"alibaba:qwen-image@2512","name":"Qwen-Image-2512","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-12-25T00:00:00Z","weight":2,"headline":"Advanced multimodal image generation model","description":"Qwen-Image-2512 is an improved version of the Qwen-Image image foundation model with enhanced prompt understanding, superior text rendering accuracy, and more realistic visual details. It generates high-fidelity images from text prompts across diverse subjects and styles.","architecture":"qwen-image","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-2512.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-2512.png","pricingMeasured":[{"configuration":"1024x1024","price":0.0051}],"pricingExamples":[{"configuration":"image-to-image · 1024×1536","price":"$0.0032","amount":0.0032,"latencyMs":7937,"exampleId":"raptor-center-annual-report-portrait"},{"configuration":"expand · 1536×1024","price":"$0.0045","amount":0.0045,"latencyMs":9126,"exampleId":"refrigerated-rail-campaign-key-visual"},{"configuration":"inpaint · 1280×1280","price":"$0.0058","amount":0.0058,"latencyMs":12051,"exampleId":"vaccine-cooler-field-lifestyle-shot"},{"configuration":"text-to-image · 1536×1024","price":"$0.0064","amount":0.0064,"latencyMs":11119,"exampleId":"pipe-organ-crew-portrait"}]},{"model":"klingai-video-2-6-pro","air":"klingai:kling-video@2.6-pro","name":"Kling VIDEO 2.6 Pro","creator":"klingai","status":"live","releasedAt":"2025-12-11T00:00:00Z","weight":2,"headline":"Kling VIDEO 2.6 Pro is a full audio-visual AI video model that combines cinematic-quality video generation with native audio (dialogue, sound effects, ambience), with optional Motion Control for precise character movement via the API.","description":"Kling VIDEO 2.6 Pro is a full audio-visual AI video model that combines cinematic-quality video generation with native audio (dialogue, sound effects, ambience). It supports flexible workflows from text or image input, delivering synchronized video and sound in one pass with strong consistency and creative control. Via the API, Motion Control enables creators to guide character movement using a reference video for more realistic and physically grounded motion.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/klingai-video-2-6-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-2-6-pro.png","pricingOverview":"From $0.07/s (without audio), or $0.14/s (with audio)","pricingRates":[{"amount":0.07,"unit":"durationSecond","label":"without audio"},{"amount":0.14,"unit":"durationSecond","label":"with audio"}],"pricingExamples":[{"configuration":"reference-to-video","price":"$0.522","amount":0.522,"latencyMs":400813,"exampleId":"indigo-dyehouse-music-video-segment"},{"configuration":"first-last-frame · 10s","price":"$0.70","amount":0.7,"latencyMs":233166,"exampleId":"archaeological-thriller-opening-title"},{"configuration":"first-frame · 5s","price":"$0.70","amount":0.7,"latencyMs":127042,"exampleId":"greenhouse-dry-break-coupler-demo"},{"configuration":"text-to-video · 1080p · 10s","price":"$1.40","amount":1.4,"latencyMs":127631,"exampleId":"heritage-seed-documentary-title-sequence"}]},{"model":"klingai-video-o1-standard","air":"klingai:kling@o1-standard","name":"Kling VIDEO O1 Standard","creator":"klingai","status":"live","releasedAt":"2025-12-01T00:00:00Z","weight":2,"headline":"Unified multimodal video generation and editing model","description":"Kling VIDEO O1 Standard is a unified multimodal video model for controllable generation and instruction-based editing. It supports text prompts, image references, and video input to enable precise control over motion, transitions, object changes, and visual adjustments within short-form video workflows.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-video-o1-standard.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-o1-standard.png","pricingRates":[{"amount":0.084,"unit":"durationSecond","label":"720p · T2V"},{"amount":0.084,"unit":"durationSecond","label":"720p · I2V"},{"amount":0.126,"unit":"durationSecond","label":"720p · R2V"}],"pricingExamples":[{"configuration":"first-last-frame · 5s","price":"$0.42","amount":0.42,"latencyMs":121105,"exampleId":"wildfire-crew-roster-update"},{"configuration":"text-to-video · 720p · 5s","price":"$0.42","amount":0.42,"latencyMs":55193,"exampleId":"archaeological-site-night-lighting-study"},{"configuration":"video-edit","price":"$0.756","amount":0.756,"latencyMs":210672,"exampleId":"rotoscoped-table-tennis-music-insert"},{"configuration":"first-frame · 10s","price":"$0.84","amount":0.84,"latencyMs":64298,"exampleId":"submarine-cable-drama-title-sequence"},{"configuration":"reference-to-video · 720p · 7s","price":"$0.882","amount":0.882,"latencyMs":321235,"exampleId":"museum-conservation-vertical-outpaint"}]},{"model":"pixverse-v5-5","air":"pixverse:1@6","name":"PixVerse V5.5","creator":"pixverse","status":"live","releasedAt":"2025-12-01T00:00:00Z","weight":2,"headline":"PixVerse V5.5 multi shot text to video with synced audio","description":"PixVerse V5.5 is a director focused video model for story driven clips. It supports multi image fusion for character continuity, multi shot sequences, and native audio. It delivers smooth motion, refined cinematic control, and precise text guided video generation for complex scenes.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/pixverse-v5-5.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v5-5.png","pricingRates":[{"amount":0.1326,"unit":"output","label":"360p · 5s"},{"amount":0.2652,"unit":"output","label":"360p · 8s"},{"amount":0.2917,"unit":"output","label":"360p · 10s"},{"amount":0.1326,"unit":"output","label":"540p · 5s"},{"amount":0.2652,"unit":"output","label":"540p · 8s"},{"amount":0.2917,"unit":"output","label":"540p · 10s"},{"amount":0.1768,"unit":"output","label":"720p · 5s"},{"amount":0.3536,"unit":"output","label":"720p · 8s"},{"amount":0.389,"unit":"output","label":"720p · 10s"},{"amount":0.3536,"unit":"output","label":"1080p · 5s"},{"amount":0.7072,"unit":"output","label":"1080p · 8s"}],"pricingExamples":[{"configuration":"first-frame · 720p · 8s","price":"$0.354","amount":0.3536,"latencyMs":96756,"exampleId":"tidal-oyster-percussion-music-segment"},{"configuration":"first-last-frame · 1080p · 5s · audio","price":"$0.383","amount":0.3831,"latencyMs":108997,"exampleId":"glass-plate-archive-title-sequence"},{"configuration":"text-to-video · 1080p · 5s · audio","price":"$0.471","amount":0.4715,"latencyMs":100261,"exampleId":"heritage-tile-restoration-reel"},{"configuration":"text-to-video · 1080p · 5s · audio","price":"$0.471","amount":0.4715,"latencyMs":114524,"exampleId":"forensic-entomology-docuseries-teaser"}]},{"model":"runway-gen-4-5","air":"runway:1@2","name":"Runway Gen-4.5","creator":"runway","status":"live","releasedAt":"2025-12-01T00:00:00Z","weight":2,"headline":"Advanced multimodal video generation with text and image input","description":"Runway Gen-4.5 is an AI video generation model that creates short video clips from text prompts or static images with high visual fidelity and smooth motion. It supports both text-to-video and image-to-video generation with a range of aspect ratios and clip durations. Gen-4.5 emphasizes realistic motion, strong prompt adherence, and controllable composition, making it suitable for cinematic sequences and creative video workflows.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/runway-gen-4-5.jpg","ogImage":"https://assets.runware.ai/og/runway-gen-4-5.png","pricingRates":[{"amount":0.12,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 5s","price":"$0.605","amount":0.605,"latencyMs":100162,"exampleId":"urban-mining-animated-logo-sting"},{"configuration":"text-to-video · 720p · 5s","price":"$0.605","amount":0.605,"latencyMs":105887,"exampleId":"ceramic-factory-music-video-sequence"},{"configuration":"first-frame · 832×1104 · 8s","price":"$0.965","amount":0.965,"latencyMs":265325,"exampleId":"pipe-inspection-crawler-demo"},{"configuration":"first-frame · 832×1104 · 8s","price":"$0.965","amount":0.965,"latencyMs":258041,"exampleId":"meteorite-documentary-title-teaser"}],"guides":[{"slug":"directing-motion","title":"Directing motion in image-to-video prompts","description":"How to write Runway Gen-4.5 image-to-video prompts that direct motion instead of redescribing the scene, using the camera and subject channels.","date":"2026-06-04","url":"https://runware.ai/docs/models/runway-gen-4-5/guides/directing-motion"}]},{"model":"alibaba-z-image-turbo","air":"runware:z-image@turbo","name":"Z-Image-Turbo","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-26T00:00:00Z","weight":2,"headline":"Fast photorealistic image generator with text control","description":"Z-Image-Turbo is a distilled vision model for sub second image generation. It produces sharp photorealistic results and supports accurate Chinese text and English text inside images. It follows complex layout instructions with stable structure for UI, posters, and scenes.","architecture":"z-image-turbo","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-z-image-turbo.jpg","ogImage":"https://assets.runware.ai/og/alibaba-z-image-turbo.png","pricingOverview":"~$0.0034 per megapixel","pricingMeasured":[{"configuration":"1024×1024","price":0.0032},{"configuration":"512×512","price":0.0013},{"configuration":"2048×2048","price":0.0141}],"pricingExamples":[{"configuration":"image-to-image · 1280×768","price":"$0.0006","amount":0.0006,"latencyMs":2325,"exampleId":"artisan-falconry-hood-lifestyle-composite"},{"configuration":"expand · 1344×768","price":"$0.0006","amount":0.0006,"latencyMs":3613,"exampleId":"cranberry-scanner-harvest-lifestyle-photograph"},{"configuration":"inpaint · 768×1152","price":"$0.0013","amount":0.0013,"latencyMs":4244,"exampleId":"rail-operations-dispatcher-avatar"},{"configuration":"text-to-image · 1024×1024","price":"$0.0013","amount":0.0013,"latencyMs":1774,"exampleId":"tidal-turbine-editorial-illustration"}]},{"model":"alibaba-qwen-image-edit-2511","air":"alibaba:qwen-image-edit@2511","name":"Qwen-Image-Edit-2511","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-25T00:00:00Z","weight":2,"headline":"Advanced instruction-driven image editing model","description":"Qwen-Image-Edit-2511 is an image editing model that applies text instructions to modify existing images with precise semantic and appearance control. It preserves visual consistency during edits, supports multi-person and character consistency, and integrates selected features and extensions that enhance object manipulation, geometric reasoning, and layout coherence.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-edit-2511.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-edit-2511.png","pricingRates":[{"amount":0.0122,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0063","amount":0.0063,"latencyMs":9854,"exampleId":"coastal-tidal-pool-night-visualization"},{"configuration":"reference-to-image","price":"$0.00714","amount":0.00714,"latencyMs":10679,"exampleId":"handheld-marine-radio-colorway"},{"configuration":"reference-to-image","price":"$0.0098","amount":0.0098,"latencyMs":14283,"exampleId":"cathedral-conservation-team-portrait"},{"configuration":"reference-to-image","price":"$0.0108","amount":0.01078,"latencyMs":15717,"exampleId":"adaptive-rain-cape-virtual-try-on"}]},{"model":"meta-sam-3d-objects","air":"meta:sam@3d","name":"SAM 3D Objects","creator":"meta","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-11-19T00:00:00Z","weight":2,"headline":"Single-image 3D reconstruction for objects and scenes","description":"SAM 3D Objects is a foundation 3D model from Meta that reconstructs full 3D shape geometry, realistic texture, and scene layout of objects from a single RGB image. It produces high-fidelity 3D meshes robustly even in cluttered, occluded real-world scenes, and significantly outperforms prior single-image 3D methods. SAM 3D Objects is part of the broader SAM 3D family, enabling 2D-to-3D reconstruction workflows for visualization, AR/VR, robotics, and digital content creation.","capabilities":["io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/meta-sam-3d-objects.jpg","ogImage":"https://assets.runware.ai/og/meta-sam-3d-objects.png","pricingMeasured":[{"configuration":"Default","price":0.0064,"latencyMs":63015}],"pricingExamples":[{"configuration":"image-to-3d","price":"$0.0051","amount":0.0051,"latencyMs":49604,"exampleId":"vampire-squid-exhibit-creature-model"},{"configuration":"image-to-3d","price":"$0.0051","amount":0.0051,"latencyMs":71440,"exampleId":"polar-clinic-set-digital-twin"},{"configuration":"image-to-3d","price":"$0.0051","amount":0.0051,"latencyMs":103258,"exampleId":"stop-motion-laundromat-set-maquette"},{"configuration":"image-to-3d","price":"$0.009","amount":0.009,"latencyMs":103869,"exampleId":"demining-detection-rat-vr-model"}]},{"model":"openai-gpt-oss-safeguard-20b","air":"openai:gpt-oss-safeguard@20b","name":"gpt-oss-safeguard-20b","creator":"openai","hosted":true,"zdr":true,"endpoints":["chat-completions"],"status":"compatible-only","releasedAt":"2025-10-29T00:00:00Z","weight":2,"headline":"Open-weight safety classifier that judges content against the policy you write in the prompt","description":"gpt-oss-safeguard-20b is OpenAI’s open-weight safety classifier. Instead of a fixed taxonomy, it takes the policy in the prompt and labels each item against it, so the same model can serve teams whose rules differ. It reasons before answering, which leaves a trace of why an item was labelled the way it was, and its 8K context window holds a policy alongside the content under review. It fits moderation pipelines and review queues where the rules change per product or per customer.","capabilities":["io:text-to-text","op:moderation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-oss-safeguard-20b.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-oss-safeguard-20b.png","contextWindow":8192,"maxOutputTokens":8192,"pricingBasis":"token","pricingRates":[{"amount":7e-8,"unit":"inputToken"},{"amount":2.8e-7,"unit":"outputToken"},{"amount":3.5e-8,"unit":"cachedInputToken"}]},{"model":"minimax-hailuo-2-3-fast","air":"minimax:4@2","name":"MiniMax Hailuo 2.3 Fast","creator":"minimax","zdr":true,"status":"live","releasedAt":"2025-10-29T00:00:00Z","weight":2,"headline":"Fast MiniMax Hailuo 2.3 model for short cinematic video","description":"MiniMax Hailuo 2.3 Fast is the speed tier of the Hailuo 2.3 video family. It targets rapid iteration for social clips, ads, and previews. It produces 6 second 768p or 1080p outputs with smooth motion and stable composition. Ideal for high volume image driven video workflows.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-hailuo-2-3-fast.jpg","ogImage":"https://assets.runware.ai/og/minimax-hailuo-2-3-fast.png","pricingOverview":"Currently 20% off for February 2026!","pricingRates":[{"amount":0.0316667,"unit":"durationSecond","label":"768p"},{"amount":0.055,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-frame","price":"$0.32","amount":0.32,"latencyMs":100865,"exampleId":"nuclear-safety-briefing-presenter"},{"configuration":"first-frame","price":"$0.32","amount":0.32,"latencyMs":131696,"exampleId":"smart-beehive-monitor-social-reel"},{"configuration":"first-frame","price":"$0.32","amount":0.32,"latencyMs":116894,"exampleId":"analog-seed-laboratory-music-video"},{"configuration":"first-frame","price":"$0.32","amount":0.32,"latencyMs":89619,"exampleId":"heritage-signal-workshop-logo-loop"}]},{"model":"minimax-hailuo-2-3","air":"minimax:4@1","name":"MiniMax Hailuo 2.3","creator":"minimax","zdr":true,"status":"live","releasedAt":"2025-10-28T00:00:00Z","weight":2,"headline":"High fidelity AI video generation from text or images","description":"MiniMax Hailuo 2.3 is a cinematic video model for short form production. It accepts text prompts or image inputs and outputs 6 or 10 second clips at 768p or 1080p. It focuses on consistent motion, strong physics, and stable scenes for ads, social content, and creative shots.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-hailuo-2-3.jpg","ogImage":"https://assets.runware.ai/og/minimax-hailuo-2-3.png","pricingOverview":"From $0.28 per generation","pricingRates":[{"amount":0.28,"unit":"output","label":"768p · 6s"},{"amount":0.56,"unit":"output","label":"768p · 10s"},{"amount":0.49,"unit":"output","label":"1080p · 6s"}],"pricingExamples":[{"configuration":"text-to-video · 1080p · 6s","price":"$0.49","amount":0.49,"latencyMs":192466,"exampleId":"volcanic-observatory-safety-presenter"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.49","amount":0.49,"latencyMs":203887,"exampleId":"marine-acoustics-logo-loop"},{"configuration":"first-frame · 1366×768 · 10s","price":"$0.56","amount":0.56,"latencyMs":157292,"exampleId":"kelp-fiber-outerwear-showcase"},{"configuration":"first-frame · 1366×768 · 10s","price":"$0.56","amount":0.56,"latencyMs":160005,"exampleId":"silkworm-cocoon-sorting-b-roll"}]},{"model":"mirelo-sfx-1-5","air":"mirelo:1@1","name":"Mirelo SFX 1.5","creator":"mirelo","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-10-22T00:00:00Z","weight":2,"headline":"Mirelo SFX 1.5 for synchronized video sound design","description":"Mirelo SFX 1.5 converts video into synchronized sound effects. It targets higher audio fidelity and wider scene coverage. It helps developers add context aware soundscapes to video pipelines with faster processing and flexible integration options.","capabilities":["io:text-to-audio","io:video-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mirelo-sfx-1-5.jpg","ogImage":"https://assets.runware.ai/og/mirelo-sfx-1-5.png","pricingOverview":"Pricing starts at $0.01/s for text to audio and video to audio ","pricingRates":[{"amount":0.00025,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-audio · 6s","price":"$0.0015","amount":0.0015,"latencyMs":2614,"exampleId":"apiary-training-narration-soundscape"},{"configuration":"text-to-audio · 6s","price":"$0.0015","amount":0.0015,"latencyMs":2542,"exampleId":"avalanche-rescue-training-audio-clip"},{"configuration":"video-edit","price":"$0.0024","amount":0.0024,"latencyMs":7709,"exampleId":"watchmaking-podcast-segment-opener"},{"configuration":"video-edit","price":"$0.0024","amount":0.0024,"latencyMs":8509,"exampleId":"herbarium-specimen-mounting-lesson-clip"}]},{"model":"hyper3d-rodin-gen-2","air":"hyper3d:rodin@gen-2","name":"Rodin Gen-2","creator":"hyper3d","status":"live","releasedAt":"2025-10-01T00:00:00Z","weight":2,"headline":"10B parameter 3D generation with quad meshes, PBR textures, and local editing","description":"Rodin Gen-2 is a 10 billion parameter 3D generation model from Hyper3D built on the BANG architecture. It creates production-ready assets from text prompts or up to 5 input images, with 4x geometry improvement over Gen-1. It supports quad-dominant and raw mesh modes, PBR and shaded materials, configurable polygon counts, T/A pose control, bounding box constraints, and output in GLB, FBX, OBJ, USDZ, and STL formats.","capabilities":["io:text-to-3d","io:image-to-3d","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/hyper3d-rodin-gen-2.jpg","ogImage":"https://assets.runware.ai/og/hyper3d-rodin-gen-2.png","pricingOverview":"Pricing starts at $0.4 per generation, if add ons ( array ) parameter includes HighPack pricing is $1.2","pricingRates":[{"amount":0.4,"unit":"output","label":"standard"},{"amount":1.2,"unit":"output","label":"HighPack"}],"pricingExamples":[{"configuration":"text-to-3d","price":"$0.40","amount":0.4,"latencyMs":120893,"exampleId":"subsea-salvage-sonar-beacon-prop"},{"configuration":"text-to-3d","price":"$0.40","amount":0.4,"latencyMs":127628,"exampleId":"countertop-koji-fermentation-crock-mockup"},{"configuration":"reference-to-3d · quality medium","price":"$1.20","amount":1.2,"latencyMs":150869,"exampleId":"museum-thylacine-reconstruction-model"},{"configuration":"reference-to-3d · quality medium","price":"$1.20","amount":1.2,"latencyMs":505995,"exampleId":"1970s-community-radio-booth-set"}],"guides":[{"slug":"image-and-text-to-3d","title":"Generating 3D models from images and text","description":"How to generate 3D models with Rodin Gen-2 from a text prompt or reference images: choosing inputs, using multi-view, prompting for 3D form, and working with the GLB output.","date":"2026-06-09","url":"https://runware.ai/docs/models/hyper3d-rodin-gen-2/guides/image-and-text-to-3d"},{"slug":"topology-and-polygon-budget","title":"Controlling mesh topology and polygon budget","description":"How to match Rodin Gen-2 output to your pipeline: Quad vs Raw topology, quality presets vs custom polygon counts, HighPack 4K textures, T/A pose for rigging, and PBR vs baked materials.","date":"2026-06-09","url":"https://runware.ai/docs/models/hyper3d-rodin-gen-2/guides/topology-and-polygon-budget"}]},{"model":"veed-fabric-1-0-hosted","air":"veed:fabric@1.0-hosted","name":"VEED Fabric 1.0","creator":"veed","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-09-26T00:00:00Z","weight":2,"headline":"Runware-hosted talking-video generation from image and audio inputs","description":"This entry is Runware's hosted deployment of VEED Fabric 1.0. It generates short speaking videos by animating a static image with synchronized speech and expressive motion, aligning lip movements, facial expressions, head gestures, and body motion to an audio input while preserving the appearance of the source image. It exists separately from the TPI-hosted VEED Fabric 1.0 model because the hosted deployment has a different API surface.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/veed-fabric-1-0-hosted.jpg","ogImage":"https://assets.runware.ai/og/veed-fabric-1-0-hosted.png","pricingRates":[{"amount":0.08,"unit":"durationSecond","label":"480p"},{"amount":0.15,"unit":"durationSecond","label":"720p"}],"pricingExamples":[{"configuration":"audio-driven","price":"$1.20","amount":1.2,"latencyMs":162512,"exampleId":"coastal-radio-logo-sting"},{"configuration":"audio-driven","price":"$1.35","amount":1.35,"latencyMs":191025,"exampleId":"arctic-seed-vault-series-teaser"},{"configuration":"audio-driven","price":"$2.10","amount":2.1,"latencyMs":307005,"exampleId":"sculptural-millinery-showcase-reel"},{"configuration":"audio-driven","price":"$3.15","amount":3.15,"latencyMs":441099,"exampleId":"pearl-diver-spoken-word-performance"}]},{"model":"pixverse-v5-fast","air":"pixverse:1@5-fast","name":"PixVerse V5 Fast","creator":"pixverse","status":"live","releasedAt":"2025-09-16T00:00:00Z","weight":2,"headline":"Fast text to video and image to video generation for rapid iteration","description":"PixVerse V5 Fast is an optimized variant of PixVerse v5 designed for faster video generation and lower latency. It supports text to video and image to video workflows while prioritizing speed and responsiveness, making it suitable for rapid iteration and preview-focused pipelines where audio, templates, and advanced controls are not required.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-v5-fast.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v5-fast.png","pricingOverview":"Pricing varies by resolution and duration, optimized for fast generation across preview and iteration workflows.","pricingRates":[{"amount":0.0943,"unit":"output","label":"360p · 5s"},{"amount":0.1856,"unit":"output","label":"360p · 8s"},{"amount":0.0943,"unit":"output","label":"540p · 5s"},{"amount":0.1856,"unit":"output","label":"540p · 8s"},{"amount":0.1238,"unit":"output","label":"720p · 5s"},{"amount":0.2475,"unit":"output","label":"720p · 8s"},{"amount":0.2475,"unit":"output","label":"1080p · 5s"},{"amount":0.495,"unit":"output","label":"1080p · 8s"}],"pricingExamples":[{"configuration":"first-frame · 720p · 8s","price":"$0.247","amount":0.2475,"latencyMs":56517,"exampleId":"ceramic-percussion-music-video-segment"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.247","amount":0.2475,"latencyMs":74080,"exampleId":"letterpress-techno-music-video-segment"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.247","amount":0.2475,"latencyMs":76961,"exampleId":"pipe-crawler-inspection-demo-clip"},{"configuration":"first-frame · 720p · 8s","price":"$0.247","amount":0.2475,"latencyMs":53932,"exampleId":"olive-oil-documentary-title-sequence"}]},{"model":"creatify-aurora-v1","air":"creatify:aurora@0","name":"Aurora v1","status":"live","releasedAt":"2025-09-10T00:00:00Z","weight":2,"headline":"High-quality audio-driven avatar video generation","description":"Aurora v1 is a multimodal avatar video generation model that creates talking-head videos from a single image and an audio input. It focuses on realistic facial animation, accurate lip synchronization, and expressive motion, producing studio-quality results for spoken or musical performances.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/creatify-aurora-v1.jpg","ogImage":"https://assets.runware.ai/og/creatify-aurora-v1.png","pricingRates":[{"amount":0.14,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven","price":"$1.12","amount":1.12,"latencyMs":67024,"exampleId":"forensic-entomology-series-teaser"},{"configuration":"audio-driven","price":"$1.26","amount":1.26,"latencyMs":94905,"exampleId":"arctic-seed-vault-series-teaser"},{"configuration":"audio-driven","price":"$1.26","amount":1.26,"latencyMs":91489,"exampleId":"historic-organ-restoration-teaser"},{"configuration":"audio-driven","price":"$1.40","amount":1.4,"latencyMs":81613,"exampleId":"shellfish-hatchery-biosecurity-presenter"}]},{"model":"creatify-aurora-v1-fast","air":"creatify:aurora@fast","name":"Aurora v1 Fast","status":"live","releasedAt":"2025-09-10T00:00:00Z","weight":2,"headline":"Fast audio-driven avatar video generation","description":"Aurora v1 Fast is an optimized variant of Aurora v1 designed for faster avatar video generation. It converts a single image and audio input into a talking-head video with synchronized lip movement and facial motion, prioritizing speed and responsiveness for rapid iteration workflows.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/creatify-aurora-v1-fast.jpg","ogImage":"https://assets.runware.ai/og/creatify-aurora-v1-fast.png","pricingOverview":"$0.07 / second","pricingRates":[{"amount":0.07,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven","price":"$0.42","amount":0.42,"latencyMs":56934,"exampleId":"seed-library-logo-sting"},{"configuration":"audio-driven","price":"$0.63","amount":0.63,"latencyMs":171691,"exampleId":"mycelium-collar-fashion-showcase-reel"},{"configuration":"audio-driven","price":"$0.63","amount":0.63,"latencyMs":93185,"exampleId":"emergency-book-conservation-social-reel"},{"configuration":"audio-driven","price":"$0.70","amount":0.7,"latencyMs":63476,"exampleId":"orchid-smuggling-docuseries-teaser"}]},{"model":"heygen-avatar-iv","air":"heygen:avatar@4","name":"HeyGen Avatar IV","creator":"heygen","zdr":true,"status":"live","releasedAt":"2025-08-27T00:00:00Z","weight":2,"headline":"AI talking avatar video generation from a single photo and script","description":"HeyGen Avatar IV is a photorealistic AI avatar generation model that creates talking videos from a single image and a script or audio input. The model synchronizes voice with facial motion, expressions, and gestures to produce lifelike avatar performances. It supports multilingual speech, realistic lip synchronization, and expressive body language, enabling scalable production of presenter-style videos without cameras, actors, or studio setups.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/heygen-avatar-iv.jpg","ogImage":"https://assets.runware.ai/og/heygen-avatar-iv.png","pricingRates":[{"amount":0.0385,"unit":"durationSecond","label":"Photo avatar"},{"amount":0.0805,"unit":"durationSecond","label":"Studio or video avatar"}],"pricingExamples":[{"configuration":"remove-background · 720p","price":"$0.209","amount":0.2087,"latencyMs":122930,"exampleId":"salvage-marketplace-logo-sting"},{"configuration":"background-replacement · 720p","price":"$0.384","amount":0.3844,"latencyMs":67054,"exampleId":"seed-vault-intake-safety-briefing"},{"configuration":"scripted-photo-avatar · 1080p","price":"$0.399","amount":0.3992,"latencyMs":200362,"exampleId":"seaweed-folk-performance-interlude"},{"configuration":"audio-driven-photo-avatar · 1080p","price":"$0.522","amount":0.5221,"latencyMs":64108,"exampleId":"industrial-spoken-word-music-verse"},{"configuration":"scripted-stock-avatar · 1080p","price":"$0.712","amount":0.7116,"latencyMs":167580,"exampleId":"roller-derby-electro-pop-bridge"},{"configuration":"audio-driven-stock-avatar · 720p","price":"$2.02","amount":2.0189,"latencyMs":256068,"exampleId":"museum-microclimate-logger-explainer"}]},{"model":"sync-lipsync-2","air":"sync:lipsync-2@1","name":"lipsync-2","creator":"sync","status":"live","releasedAt":"2025-08-20T00:00:00Z","weight":2,"headline":"Zero-shot audio-driven lip synchronization for video","description":"lipsync-2 is a zero-shot lip synchronization model that aligns spoken audio to existing video while preserving the speaker’s identity and natural speaking style. It works across live-action, animation, and AI-generated footage without training or fine-tuning.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sync-lipsync-2.jpg","ogImage":"https://assets.runware.ai/og/sync-lipsync-2.png","pricingOverview":"$0.0440 per second","pricingRates":[{"amount":0.044,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"segmented-lip-sync","price":"$0.266","amount":0.265848,"latencyMs":304925,"exampleId":"seed-vault-intake-conversation-b-roll"},{"configuration":"segmented-lip-sync","price":"$0.266","amount":0.265848,"latencyMs":114991,"exampleId":"mushroom-farm-duet-music-segment"},{"configuration":"audio-driven","price":"$0.428","amount":0.42768,"latencyMs":95161,"exampleId":"nocturnal-auctioneer-spoken-word-music-video"},{"configuration":"audio-driven","price":"$0.517","amount":0.51744,"latencyMs":100342,"exampleId":"polar-station-safety-presenter"}]},{"model":"sync-lipsync-2-pro","air":"sync:lipsync-2-pro@1","name":"lipsync-2-pro","creator":"sync","status":"live","releasedAt":"2025-08-20T00:00:00Z","weight":2,"headline":"High-resolution diffusion-based lip-sync editing up to 4K","description":"lipsync-2-pro extends lipsync-2 with diffusion-based enhancement for studio-grade lip synchronization. It preserves fine facial details such as teeth, facial hair, and micro-expressions while supporting high-resolution output suitable for professional post-production workflows.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/sync-lipsync-2-pro.jpg","ogImage":"https://assets.runware.ai/og/sync-lipsync-2-pro.png","pricingOverview":"$0.0733 per second","pricingRates":[{"amount":0.0733,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"segmented-lip-sync","price":"$0.37","amount":0.36955419,"latencyMs":129042,"exampleId":"conflict-mediation-sonic-logo-sting"},{"configuration":"segmented-lip-sync","price":"$0.443","amount":0.4428786,"latencyMs":322149,"exampleId":"weather-station-spoken-bridge-segment"},{"configuration":"audio-driven","price":"$0.611","amount":0.61083331,"latencyMs":78252,"exampleId":"hearing-jewelry-showcase-reel"},{"configuration":"audio-driven","price":"$0.69","amount":0.69024169,"latencyMs":188751,"exampleId":"mycology-cleanroom-training-presenter"}]},{"model":"openai-gpt-5-mini","air":"openai:gpt@5-mini","name":"GPT-5 Mini","creator":"openai","endpoints":["chat-completions","responses"],"status":"deprecated","replacedBy":"openai-gpt-5-4-mini","deactivatesAt":"2026-12-11T00:00:00Z","releasedAt":"2025-08-07T00:00:00Z","weight":2,"headline":"Cost-efficient reasoning LLM for low-latency, high-volume workloads","description":"GPT-5 Mini is a faster, more cost-efficient variant of GPT-5 designed for well-defined tasks, precise prompts, and high-volume production use. OpenAI positions it for cost-sensitive, low-latency workloads while retaining strong reasoning ability, 400K context, 128K max output tokens, image input support, and compatibility with structured outputs and function calling.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-mini.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-mini.png","contextWindow":400000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":2.5e-7,"unit":"inputToken"},{"amount":0.000002,"unit":"outputToken"},{"amount":2.5e-8,"unit":"cachedInputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000366","amount":0.000366,"latencyMs":3377,"exampleId":"pharma-cargo-cutoff-decision"},{"configuration":"structured-output","price":"$0.0083","amount":0.008302,"latencyMs":33894,"exampleId":"commercial-lease-abstract-dataset"},{"configuration":"text-generation","price":"$0.0135","amount":0.013487,"latencyMs":94807,"exampleId":"documentary-distribution-deal-memo"},{"configuration":"reference-to-text","price":"$0.0147","amount":0.014713,"latencyMs":74931,"exampleId":"wetland-trail-camera-survey-report"}]},{"model":"openai-gpt-5-nano","air":"openai:gpt@5-nano","name":"GPT-5 Nano","creator":"openai","endpoints":["chat-completions","responses"],"status":"deprecated","replacedBy":"openai-gpt-6-luna","deactivatesAt":"2026-12-11T00:00:00Z","releasedAt":"2025-08-07T00:00:00Z","weight":2,"headline":"Fastest, lowest-cost GPT-5 variant for summarization, classification, and lightweight automation","description":"GPT-5 Nano is OpenAI's fastest and cheapest GPT-5 variant, built for speed- and cost-sensitive workloads such as summarization, classification, extraction, and lightweight automation. It supports 400K context, 128K max output tokens, image input, reasoning token support, structured outputs, and function calling in a compact package optimized for high-throughput API use.","capabilities":["io:text-to-text","io:image-to-text","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-gpt-5-nano.jpg","ogImage":"https://assets.runware.ai/og/openai-gpt-5-nano.png","contextWindow":400000,"maxOutputTokens":128000,"pricingBasis":"token","pricingRates":[{"amount":5e-8,"unit":"inputToken"},{"amount":4e-7,"unit":"outputToken"},{"amount":5e-9,"unit":"cachedInputToken"}],"pricingExamples":[{"configuration":"tool-calling","price":"$0.000022","amount":0.000022,"latencyMs":1208,"exampleId":"live-incident-status-update"},{"configuration":"reference-to-text","price":"$0.000503","amount":0.000503,"latencyMs":10182,"exampleId":"grocery-shelf-availability-audit"},{"configuration":"text-generation","price":"$0.000737","amount":0.000737,"latencyMs":10556,"exampleId":"mobile-dental-clinic-grant-summary"},{"configuration":"structured-output","price":"$0.00331","amount":0.00331,"latencyMs":42827,"exampleId":"retail-lease-risk-register"}]},{"model":"alibaba-qwen-image","air":"runware:108@1","name":"Qwen-Image","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-08-04T00:00:00Z","weight":2,"headline":"Qwen-Image high fidelity text aware image generation model","description":"Qwen-Image is a 20B parameter vision language model from Alibaba Cloud. It focuses on precise text conditioned image generation and supports complex Chinese or English typography. It also enables accurate image editing workflows that need layout control and strong prompt following.","architecture":"qwen-image","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image.png","pricingOverview":"Each image generation costs $0.0058 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 20 steps","price":0.0058}],"pricingExamples":[{"configuration":"image-to-image · 1280×768","price":"$0.0026","amount":0.0026,"latencyMs":5723,"exampleId":"pediatric-lab-journey-learning-panel"},{"configuration":"expand · 1024×768","price":"$0.0026","amount":0.0026,"latencyMs":6387,"exampleId":"meteorological-observatory-linocut-wall-art"},{"configuration":"inpaint · 768×1024","price":"$0.0026","amount":0.0026,"latencyMs":6739,"exampleId":"coastal-eco-thriller-series-poster"},{"configuration":"text-to-image · 1024×1024","price":"$0.0032","amount":0.0032,"latencyMs":5291,"exampleId":"subsea-documentary-square-expansion"}]},{"model":"pixverse-lipsync","air":"pixverse:lipsync@1","name":"PixVerse LipSync","creator":"pixverse","status":"live","releasedAt":"2025-07-30T00:00:00Z","weight":2,"headline":"Realistic AI lip sync from audio for any video","description":"PixVerse LipSync generates accurate mouth motion from audio for characters and videos. It aligns lip movement with speech timing. It preserves facial expression context. Ideal for dubbing, character animation, and content localization workflows.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-lipsync.jpg","ogImage":"https://assets.runware.ai/og/pixverse-lipsync.png","pricingOverview":"$0.0136 per second of audio","pricingRates":[{"amount":0.0136,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"scripted-lip-sync","price":"$0.0546","amount":0.0546,"latencyMs":32738,"exampleId":"community-radio-vocal-logo-loop"},{"configuration":"audio-driven","price":"$0.109","amount":0.1091,"latencyMs":63077,"exampleId":"arctic-eco-thriller-teaser-shot"},{"configuration":"scripted-lip-sync","price":"$0.123","amount":0.1228,"latencyMs":49959,"exampleId":"manuscript-scanner-product-explainer"},{"configuration":"audio-driven","price":"$0.136","amount":0.1364,"latencyMs":57073,"exampleId":"oyster-hatchery-sensor-explainer"}]},{"model":"minimax-hailuo-02","air":"minimax:3@1","name":"MiniMax Hailuo 02","creator":"minimax","zdr":true,"status":"live","releasedAt":"2025-06-19T00:00:00Z","weight":2,"headline":"Cinematic AI video model for viral and commercial clips","description":"MiniMax Hailuo 02 is a 1080p AI video model for cinematic, high motion scenes. It converts text prompts or still images into short, polished clips with strong instruction following and realistic physics. Ideal for commercial spots, trailers, music promos, and social shorts.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-hailuo-02.jpg","ogImage":"https://assets.runware.ai/og/minimax-hailuo-02.png","pricingOverview":"Each generation will cost $0.0560/s for 768p, or $0.0817/s for 1080p.","pricingRates":[{"amount":0.056,"unit":"durationSecond","label":"768p"},{"amount":0.0816667,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 1366×768 · 6s","price":"$0.28","amount":0.28,"latencyMs":108172,"exampleId":"historic-stone-laser-cleaning-reel"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.49","amount":0.49,"latencyMs":172355,"exampleId":"forensic-botany-docuseries-title-teaser"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.49","amount":0.49,"latencyMs":154266,"exampleId":"sculptural-hearing-jewelry-fashion-reel"},{"configuration":"first-frame · 1366×768 · 10s","price":"$0.56","amount":0.56,"latencyMs":294674,"exampleId":"couture-beekeeping-suit-showcase-reel"}]},{"model":"stabilityai-stable-diffusion-3","air":"runware:5@1","name":"Stable Diffusion 3","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-06-12T00:00:00Z","weight":2,"headline":"Stable Diffusion 3 for sharper text and complex images","description":"Stable Diffusion 3 is a next generation text to image model with improved prompt adherence and typography. It handles complex scenes with multiple subjects and fine detail. It targets both local and cloud deployment so developers can integrate high quality image generation into products.","architecture":"sd3","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/stabilityai-stable-diffusion-3.jpg","ogImage":"https://assets.runware.ai/og/stabilityai-stable-diffusion-3.png","pricingOverview":"~$0.0043 per megapixel","pricingMeasured":[{"configuration":"512×512 · steps 28","price":0.0006,"latencyMs":2341},{"configuration":"1024×1024 · steps 28","price":0.0013,"latencyMs":3977},{"configuration":"1024×1024 · steps 50","price":0.0026,"latencyMs":4994},{"configuration":"2048×2048 · steps 28","price":0.007,"latencyMs":13190}],"pricingExamples":[{"configuration":"image-to-image · 1280×768","price":"$0.0013","amount":0.0013,"latencyMs":10171,"exampleId":"electric-snowplow-fleet-blog-hero"},{"configuration":"text-to-image · 1024×1024","price":"$0.0013","amount":0.0013,"latencyMs":10172,"exampleId":"marina-biofouling-inspection-guide-illustration"},{"configuration":"inpaint · 1152×896","price":"$0.0013","amount":0.0013,"latencyMs":10993,"exampleId":"museum-conservation-paste-tin-label"},{"configuration":"control-guided · 1024×1536","price":"$0.0019","amount":0.0019,"latencyMs":11622,"exampleId":"racing-wheelchair-colorway-visualization"},{"configuration":"expand · 1536×1024","price":"$0.0026","amount":0.0026,"latencyMs":14231,"exampleId":"mushroom-farm-biosecurity-sticker-pack"}]},{"model":"pixverse-v4-5","air":"pixverse:1@3","name":"PixVerse V4.5","creator":"pixverse","status":"live","releasedAt":"2025-05-14T00:00:00Z","weight":2,"headline":"PixVerse V4.5 cinematic text and image to video model","description":"PixVerse V4.5 generates stylized cinematic video from text prompts or reference images. It adds refined camera motion control, multi image fusion, and faster modes for iteration. Ideal for creators who need dynamic shots, complex motion, and consistent stylized outputs.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-v4-5.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v4-5.png","pricingOverview":"Each generation will cost $0.0598/s for 1080p.","pricingRates":[{"amount":0.0224,"unit":"durationSecond","label":"360p"},{"amount":0.0224,"unit":"durationSecond","label":"540p"},{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.0598,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"visual-effect · 540p · 5s","price":"$0.112","amount":0.112,"latencyMs":73829,"exampleId":"museum-after-dark-logo-sting"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.299","amount":0.299,"latencyMs":42054,"exampleId":"technical-apiary-suit-fashion-reel"},{"configuration":"first-frame · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":54538,"exampleId":"subsea-tidal-turbine-inspection-b-roll"},{"configuration":"text-to-video · 720p · 5s","price":"$0.299","amount":0.299,"latencyMs":46058,"exampleId":"fish-market-percussion-music-video"}]},{"model":"controlnet-preprocess-canny","air":"runware:controlnet-preprocess@canny","name":"ControlNet Preprocess Canny","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Edge extraction preprocessor for strong shape-preserving ControlNet guidance","description":"ControlNet Preprocess Canny converts an input image into a high-contrast Canny edge map for ControlNet conditioning. It is best suited to workflows that need clear object boundaries, strong silhouette retention, and tightly constrained structure during guided image generation or editing.","capabilities":["io:image-to-image","op:preprocess","op:edge-detection","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-canny.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-canny.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":4116}],"pricingExamples":[{"configuration":"canny-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":5351,"exampleId":"roman-mosaic-cleanup-edge-guide"},{"configuration":"canny-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":4692,"exampleId":"saturation-diver-headshot-edge-guide"},{"configuration":"canny-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":3236,"exampleId":"watchmaker-team-composite-edge-guide"},{"configuration":"canny-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":3139,"exampleId":"wheelchair-racer-arena-composite"}]},{"model":"controlnet-preprocess-depth","air":"runware:controlnet-preprocess@depth","name":"ControlNet Preprocess Depth","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Monocular depth-map preprocessor for scene geometry and spatial consistency","description":"ControlNet Preprocess Depth estimates a depth map from an input image for ControlNet conditioning. It is useful when the goal is to preserve scene layout, foreground-background separation, camera perspective, and overall 3D spatial structure while changing style or content.","capabilities":["io:image-to-image","op:preprocess","op:depth-estimation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-depth.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-depth.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":3431}],"pricingExamples":[{"configuration":"depth-estimation","price":"$0.0013","amount":0.0013,"latencyMs":4222,"exampleId":"forensic-entomologist-portrait-depth-guide"},{"configuration":"depth-estimation","price":"$0.0013","amount":0.0013,"latencyMs":6113,"exampleId":"polar-night-plant-depth-guide"},{"configuration":"depth-estimation","price":"$0.0013","amount":0.0013,"latencyMs":4941,"exampleId":"hatchery-brochure-cleanup-depth-guide"},{"configuration":"depth-estimation","price":"$0.0019","amount":0.0019,"latencyMs":5921,"exampleId":"diving-suit-background-replacement-guide"}]},{"model":"controlnet-preprocess-lineart","air":"runware:controlnet-preprocess@lineart","name":"ControlNet Preprocess Lineart","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Clean contour extraction preprocessor for illustration and stylized line-driven control","description":"ControlNet Preprocess Lineart extracts simplified line art from an input image for ControlNet conditioning. It is useful when the goal is to preserve contours, composition, and object shapes while translating the result into illustration, concept art, or other stylized outputs.","capabilities":["io:image-to-image","op:preprocess","op:line-extraction","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-lineart.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-lineart.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3996}],"pricingExamples":[{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":5200,"exampleId":"avalanche-patrol-portrait-line-art-guide"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3897,"exampleId":"arborist-helmet-colorway-line-guide"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3829,"exampleId":"deep-sea-exhibit-panorama-guide"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3992,"exampleId":"hydroelectric-mural-expansion-guide"}]},{"model":"controlnet-preprocess-lineart-anime","air":"runware:controlnet-preprocess@lineart-anime","name":"ControlNet Preprocess Lineart Anime","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Anime-focused line extraction preprocessor for cel-style and manga-oriented guidance","description":"ControlNet Preprocess Lineart Anime extracts anime-style line structure from an input image for ControlNet conditioning. It is best for manga, cel-shaded, and stylized character workflows where cleaner region boundaries and illustration-style contours matter more than photoreal edge detail.","capabilities":["io:image-to-image","op:preprocess","op:line-extraction","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-lineart-anime.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-lineart-anime.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":4519}],"pricingExamples":[{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3963,"exampleId":"mine-safety-canary-logo-vectorization"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":5916,"exampleId":"heritage-tram-coloring-page"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3900,"exampleId":"antarctic-crew-roster-line-guide"},{"configuration":"lineart-extraction","price":"$0.0006","amount":0.0006,"latencyMs":5824,"exampleId":"aquarium-mural-outpaint-line-guide"}]},{"model":"controlnet-preprocess-mlsd","air":"runware:controlnet-preprocess@mlsd","name":"ControlNet Preprocess MLSD","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Straight-line detection preprocessor for architecture, interiors, and clean geometry","description":"ControlNet Preprocess MLSD extracts major straight line segments from an input image for ControlNet conditioning. It is especially effective for buildings, interiors, product shots, and other compositions where rigid linear structure matters more than texture or organic detail.","capabilities":["io:image-to-image","op:preprocess","op:line-extraction","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-mlsd.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-mlsd.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3159}],"pricingExamples":[{"configuration":"line-segment-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3979,"exampleId":"pipe-organ-venue-swap-guide"},{"configuration":"line-segment-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3457,"exampleId":"seed-vault-emblem-vector-trace"},{"configuration":"line-segment-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3093,"exampleId":"modular-sauna-background-replacement-guide"},{"configuration":"line-segment-extraction","price":"$0.0013","amount":0.0013,"latencyMs":4307,"exampleId":"sento-gouache-restyle-structure-map"}]},{"model":"controlnet-preprocess-normalbae","air":"runware:controlnet-preprocess@normalbae","name":"ControlNet Preprocess NormalBae","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Surface-normal preprocessor for form, volume, and lighting-aware structure control","description":"ControlNet Preprocess NormalBae predicts a surface normal map from an input image for ControlNet conditioning. It is well suited to workflows that need stronger preservation of 3D form, object volume, and orientation cues beyond what a flat edge map or line map can provide.","capabilities":["io:image-to-image","op:preprocess","op:surface-normal-estimation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-normalbae.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-normalbae.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":5104}],"pricingExamples":[{"configuration":"surface-normal-estimation","price":"$0.0006","amount":0.0006,"latencyMs":3947,"exampleId":"marionette-photograph-restoration-normal-map"},{"configuration":"surface-normal-estimation","price":"$0.0013","amount":0.0013,"latencyMs":6941,"exampleId":"aviation-archive-restoration-normal-pass"},{"configuration":"surface-normal-estimation","price":"$0.0013","amount":0.0013,"latencyMs":5880,"exampleId":"underwater-rov-compositing-normal-pass"},{"configuration":"surface-normal-estimation","price":"$0.0013","amount":0.0013,"latencyMs":4846,"exampleId":"pipe-organ-panorama-normal-guide"}]},{"model":"controlnet-preprocess-openpose","air":"runware:controlnet-preprocess@openpose","name":"ControlNet Preprocess OpenPose","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Human pose preprocessor for body, hand, and facial landmark-driven guidance","description":"ControlNet Preprocess OpenPose extracts a human pose skeleton from an input image for ControlNet conditioning. It is best for character and figure workflows where body posture, gesture, limb placement, and overall pose need to stay consistent across generations.","capabilities":["io:image-to-image","op:preprocess","op:pose-estimation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-openpose.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-openpose.png","pricingMeasured":[{"configuration":"Body pose only (default)","price":0.0026,"latencyMs":6168}],"pricingExamples":[{"configuration":"pose-estimation","price":"$0.0026","amount":0.0026,"latencyMs":6624,"exampleId":"volcanologist-wpa-poster-restyle"},{"configuration":"detailed-pose-estimation","price":"$0.0032","amount":0.0032,"latencyMs":7674,"exampleId":"roller-derby-roster-addition-pose-map"},{"configuration":"pose-estimation","price":"$0.0032","amount":0.0032,"latencyMs":7478,"exampleId":"motorcycle-airbag-suit-pose-map"},{"configuration":"detailed-pose-estimation","price":"$0.0038","amount":0.0038,"latencyMs":9535,"exampleId":"museum-interpreter-scene-swap-pose-map"}]},{"model":"controlnet-preprocess-scribble","air":"runware:controlnet-preprocess@scribble","name":"ControlNet Preprocess Scribble","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Sparse sketch preprocessor for loose composition and guided redraw workflows","description":"ControlNet Preprocess Scribble converts an input image into a simplified scribble-style guide for ControlNet conditioning. It is best for workflows that want loose structural guidance, rough composition transfer, or sketch-based reinterpretation instead of strict edge preservation.","capabilities":["io:image-to-image","op:preprocess","op:line-extraction","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-scribble.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-scribble.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3418}],"pricingExamples":[{"configuration":"scribble-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3446,"exampleId":"damaged-clinic-bus-restoration-guide"},{"configuration":"scribble-extraction","price":"$0.0006","amount":0.0006,"latencyMs":6685,"exampleId":"avalanche-team-portrait-scribble-guide"},{"configuration":"scribble-extraction","price":"$0.0006","amount":0.0006,"latencyMs":4218,"exampleId":"aerospace-founder-headshot-retouch-guide"},{"configuration":"scribble-extraction","price":"$0.0006","amount":0.0006,"latencyMs":3222,"exampleId":"bathysphere-panorama-expansion-guide"}]},{"model":"controlnet-preprocess-seg","air":"runware:controlnet-preprocess@seg","name":"ControlNet Preprocess Seg","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Semantic segmentation preprocessor for region-aware ControlNet composition control","description":"ControlNet Preprocess Seg converts an input image into a semantic segmentation map for ControlNet conditioning. It is best for preserving broad scene organization such as sky, ground, buildings, clothing, vegetation, and other region-level structure while allowing major stylistic changes.","capabilities":["io:image-to-image","op:preprocess","op:segmentation","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-seg.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-seg.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":4450}],"pricingExamples":[{"configuration":"semantic-segmentation","price":"$0.0013","amount":0.0013,"latencyMs":4881,"exampleId":"oyster-hatchery-dusk-relight-guide"},{"configuration":"semantic-segmentation","price":"$0.0013","amount":0.0013,"latencyMs":6546,"exampleId":"glassblower-headshot-retouch-segmentation"},{"configuration":"semantic-segmentation","price":"$0.0013","amount":0.0013,"latencyMs":5839,"exampleId":"cargo-trike-colorway-segmentation-guide"},{"configuration":"semantic-segmentation","price":"$0.0013","amount":0.0013,"latencyMs":4557,"exampleId":"drysuit-colorway-segmentation-map"}]},{"model":"controlnet-preprocess-shuffle","air":"runware:controlnet-preprocess@shuffle","name":"ControlNet Preprocess Shuffle","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Content-shuffle preprocessor for color and texture transfer with looser spatial control","description":"ControlNet Preprocess Shuffle creates a shuffled representation of the source image for ControlNet conditioning. It is useful when the goal is to borrow palette, local texture, and visual character from a source image without rigidly preserving exact edges, contours, or pose.","capabilities":["io:image-to-image","op:preprocess","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-shuffle.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-shuffle.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":2867}],"pricingExamples":[{"configuration":"content-shuffle","price":"$0.0006","amount":0.0006,"latencyMs":3061,"exampleId":"polar-night-station-relight-guide"},{"configuration":"content-shuffle","price":"$0.0006","amount":0.0006,"latencyMs":3368,"exampleId":"skate-championship-broadcast-expansion-guide"},{"configuration":"content-shuffle","price":"$0.0006","amount":0.0006,"latencyMs":4716,"exampleId":"night-market-cooktop-scene-guide"},{"configuration":"content-shuffle","price":"$0.0006","amount":0.0006,"latencyMs":6290,"exampleId":"geology-exhibition-panorama-expansion"}]},{"model":"controlnet-preprocess-softedge","air":"runware:controlnet-preprocess@softedge","name":"ControlNet Preprocess SoftEdge","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Soft-edge preprocessor for gentler contour guidance than hard Canny boundaries","description":"ControlNet Preprocess SoftEdge extracts smoother, more organic edge structure from an input image for ControlNet conditioning. It is useful when the goal is to preserve contours and layout without the harsh binary boundaries of Canny, especially for painterly, naturalistic, or softer illustration workflows.","capabilities":["io:image-to-image","op:preprocess","op:edge-detection","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-softedge.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-softedge.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3052}],"pricingExamples":[{"configuration":"soft-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":4327,"exampleId":"harbor-pilot-headshot-edge-guide"},{"configuration":"soft-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":4300,"exampleId":"espresso-machine-colorway-edge-guide"},{"configuration":"soft-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":4489,"exampleId":"alpine-tram-heritage-lithograph"},{"configuration":"soft-edge-detection","price":"$0.0006","amount":0.0006,"latencyMs":3478,"exampleId":"vintage-synthesizer-catalog-cleanup-guide"}]},{"model":"controlnet-preprocess-tile","air":"runware:controlnet-preprocess@tile","name":"ControlNet Preprocess Tile","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":2,"headline":"Tile-based preprocessor for detail reconstruction, enhancement, and local consistency","description":"ControlNet Preprocess Tile prepares an image for tile-style ControlNet conditioning, emphasizing local detail and patch-level consistency. It is useful for upscaling-adjacent workflows, texture refinement, and preserving fine local structure when a model needs more detailed guidance from the source image.","capabilities":["io:image-to-image","op:preprocess","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/controlnet-preprocess-tile.jpg","ogImage":"https://assets.runware.ai/og/controlnet-preprocess-tile.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":4216}],"pricingExamples":[{"configuration":"tile-preprocessing","price":"$0.0006","amount":0.0006,"latencyMs":5670,"exampleId":"wildland-ecologist-headshot-detail-map"},{"configuration":"tile-preprocessing","price":"$0.0006","amount":0.0006,"latencyMs":3855,"exampleId":"woodcut-organ-restyle-detail-guide"},{"configuration":"tile-preprocessing","price":"$0.0006","amount":0.0006,"latencyMs":4179,"exampleId":"historic-planetarium-restoration-control-map"},{"configuration":"tile-preprocessing","price":"$0.0006","amount":0.0006,"latencyMs":4680,"exampleId":"accordion-colorway-detail-guide"}]},{"model":"vidu-q1","air":"vidu:1@1","name":"Vidu Q1","creator":"vidu","status":"live","releasedAt":"2025-04-21T00:00:00Z","weight":2,"headline":"Vidu Q1 high fidelity reference to video generation model","description":"Vidu Q1 is a generative video model that preserves visual fidelity from multiple reference images. It supports character, scene and prop control with smooth transitions and 1080p clips. Ideal for ads, story sequences and animation workflows that need tight visual continuity.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-q1.jpg","coverVideo":"https://assets.runware.ai/34593ec9-7469-4283-a657-6d9bad03145b.mp4","ogImage":"https://assets.runware.ai/og/vidu-q1.png","pricingOverview":"Each generation will cost $0.0440/s for 1080p.","pricingRates":[{"amount":0.22,"unit":"output","label":"1080p"}],"pricingExamples":[{"configuration":"text-to-video","price":"$0.22","amount":0.22,"latencyMs":301870,"exampleId":"volcanic-observatory-anime-series-teaser"},{"configuration":"first-frame","price":"$0.22","amount":0.22,"latencyMs":135014,"exampleId":"magnetic-silo-crawler-demo-clip"},{"configuration":"first-last-frame","price":"$0.22","amount":0.22,"latencyMs":113659,"exampleId":"cuneiform-docuseries-title-sequence"},{"configuration":"text-to-video","price":"$0.22","amount":0.22,"latencyMs":244776,"exampleId":"apiary-workwear-fashion-reel"}]},{"model":"vidu-q1-classic","air":"vidu:1@0","name":"Vidu Q1 Classic","creator":"vidu","status":"live","releasedAt":"2025-04-21T00:00:00Z","weight":2,"headline":"Vidu Q1 Classic 1080p Text and Image to Video Generation","description":"Vidu Q1 Classic generates 1080p clips up to 16 seconds from text prompts, source images, or reference shots. It targets controllable motion and stable scenes for fast prototyping. Ideal for teams that need cinematic tests without complex video pipelines.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-q1-classic.jpg","ogImage":"https://assets.runware.ai/og/vidu-q1-classic.png","pricingRates":[{"amount":0.22,"unit":"output","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame","price":"$0.22","amount":0.22,"latencyMs":123537,"exampleId":"polar-archive-exhibition-title-sequence"},{"configuration":"first-frame","price":"$0.22","amount":0.22,"latencyMs":107439,"exampleId":"wildfire-investigation-docuseries-title-teaser"},{"configuration":"first-last-frame","price":"$0.22","amount":0.22,"latencyMs":157786,"exampleId":"recycled-glass-logo-sting"},{"configuration":"first-frame","price":"$0.22","amount":0.22,"latencyMs":135747,"exampleId":"marine-oil-skimmer-demonstration-clip"}]},{"model":"minimax-01-live","air":"minimax:2@3","name":"MiniMax 01 Live","creator":"minimax","zdr":true,"status":"live","releasedAt":"2025-04-14T00:00:00Z","weight":2,"headline":"Anime video model for expressive character animation","description":"MiniMax 01 Live generates short stylized videos from static anime art. It focuses on expressive character motion with consistent details. Use it to turn illustrations or manga panels into dynamic clips suitable for cutscenes, social posts, or prototype shots.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-01-live.jpg","ogImage":"https://assets.runware.ai/og/minimax-01-live.png","pricingOverview":"Each generation will cost $0.0560/s for 720p.","pricingRates":[{"amount":0.056,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"first-frame","price":"$0.28","amount":0.28,"latencyMs":369186,"exampleId":"rare-manuscript-conservation-b-roll"},{"configuration":"first-frame · 6s","price":"$0.28","amount":0.28,"latencyMs":216304,"exampleId":"climate-documentary-series-title-sequence"},{"configuration":"first-frame · 6s","price":"$0.28","amount":0.28,"latencyMs":167662,"exampleId":"regenerative-kelp-synth-pop-video"},{"configuration":"first-frame","price":"$0.28","amount":0.28,"latencyMs":418569,"exampleId":"semiconductor-cleanroom-couture-reel"}]},{"model":"hidream-i1-dev","air":"runware:97@2","name":"HiDream-I1 Dev","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-04-08T00:00:00Z","weight":2,"headline":"HiDream-I1 Dev fast 17B text to image generation model","description":"HiDream-I1 Dev is a distilled 17B text to image model that balances speed and quality. It runs in about 28 diffusion steps and supports LoRAs for style control. Ideal for rapid iteration, style exploration, and clean concept rendering in production workflows.","architecture":"hidream-dev","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/hidream-i1-dev.jpg","ogImage":"https://assets.runware.ai/og/hidream-i1-dev.png","pricingOverview":"Each image generation costs $0.0045 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 28 steps","price":0.0045}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0026","amount":0.0026,"latencyMs":4613,"exampleId":"subsea-infrastructure-magazine-cover"},{"configuration":"text-to-image","price":"$0.0026","amount":0.0026,"latencyMs":4536,"exampleId":"stoneware-fermentation-crock-product-packshot"},{"configuration":"text-to-image","price":"$0.0026","amount":0.0026,"latencyMs":5588,"exampleId":"childrens-paleontology-host-character-portrait"},{"configuration":"text-to-image","price":"$0.0026","amount":0.0026,"latencyMs":4506,"exampleId":"high-rise-cleaner-personal-portrait"}]},{"model":"hidream-i1-full","air":"runware:97@1","name":"HiDream-I1 Full","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-04-07T00:00:00Z","weight":2,"headline":"HiDream-I1 Full high fidelity text to image generator","description":"HiDream-I1 Full is a 17B parameter text to image model for high quality generation. It targets sharp detail and strong prompt alignment. It supports LoRA workflows and heavy customization. Ideal for production pipelines that need consistent visuals and open source licensing.","architecture":"hidream-full","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/hidream-i1-full.jpg","ogImage":"https://assets.runware.ai/og/hidream-i1-full.png","pricingOverview":"Each image generation costs $0.009 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 30 steps","price":0.009}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0045","amount":0.0045,"latencyMs":10095,"exampleId":"illegal-timber-forensics-blog-hero"},{"configuration":"text-to-image","price":"$0.0045","amount":0.0045,"latencyMs":10003,"exampleId":"subsea-cable-incident-reconstruction"},{"configuration":"text-to-image","price":"$0.0045","amount":0.0045,"latencyMs":9154,"exampleId":"premium-poultry-waterer-packshot"},{"configuration":"text-to-image","price":"$0.0051","amount":0.0051,"latencyMs":8649,"exampleId":"wildland-fire-strategy-game-key-art"}]},{"model":"qwen2-5-vl-7b-age-detector","air":"runware:152@50","name":"Qwen2.5-VL-7B Age Detector","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-04-06T00:00:00Z","weight":2,"headline":"Vision-language age estimation model","description":"Qwen2.5-VL-7B Age Detector is a multimodal model that analyzes a facial image to estimate age. It uses the vision encoder from Qwen2.5-VL-7B and its instruction-tuned language backbone to interpret visual features and output age predictions or age categories.","capabilities":["io:image-to-text","op:age-classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/qwen2-5-vl-7b-age-detector.jpg","ogImage":"https://assets.runware.ai/og/qwen2-5-vl-7b-age-detector.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5147}],"pricingExamples":[{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":4649,"exampleId":"insurance-spokesperson-casting-age-review"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":13150,"exampleId":"alcohol-delivery-age-triage"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":4693,"exampleId":"clinical-trial-age-cohort-review"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":4566,"exampleId":"culinary-talent-age-band-assessment"}]},{"model":"vidu-q1-image","air":"vidu:q1@image","name":"Vidu Q1","creator":"vidu","status":"live","releasedAt":"2025-03-29T00:00:00Z","weight":2,"headline":"High-fidelity multi-reference image synthesis","description":"Vidu Q1 (image) is a reference-to-image model designed for high visual fidelity. It blends multiple input images with consistent identity and style. Prompts can guide composition and layout without losing coherence. The model supports flexible aspect ratios for ads, social content, storyboards or animation assets. It produces clean visuals with minimal effort and is useful for rapid creative workflows.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-q1-image.jpg","ogImage":"https://assets.runware.ai/og/vidu-q1-image.png","pricingRates":[{"amount":0.055,"unit":"output","label":"1080p"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.055","amount":0.055,"latencyMs":35184,"exampleId":"master-horologist-press-headshot"},{"configuration":"reference-to-image","price":"$0.055","amount":0.055,"latencyMs":40479,"exampleId":"subterranean-survey-crew-portrait"},{"configuration":"reference-to-image","price":"$0.055","amount":0.055,"latencyMs":47497,"exampleId":"tidal-forager-workwear-lookbook"},{"configuration":"reference-to-image","price":"$0.055","amount":0.055,"latencyMs":37625,"exampleId":"estonian-bog-sauna-tourism-campaign"}]},{"model":"ideogram-3-0","air":"ideogram:4@1","name":"Ideogram 3.0","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-03-26T00:00:00Z","weight":2,"headline":"Ideogram 3.0 text to image model for sharp design visuals","description":"Ideogram 3.0 is a text to image model for high fidelity design work. It improves text rendering, complex layout handling, and photorealism. It also adds stronger style controls and supports editing tasks like inpainting and background replacement for production workflows.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-3-0.jpg","ogImage":"https://assets.runware.ai/og/ideogram-3-0.png","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.09,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"identity-preservation · 896×1152","price":"$0.09","amount":0.09,"latencyMs":94740,"exampleId":"geothermal-careers-recruitment-campaign-visual"},{"configuration":"text-to-image · 1024×1024","price":"$0.09","amount":0.09,"latencyMs":26037,"exampleId":"ceramic-glaze-powder-pouch-label"},{"configuration":"text-to-image · 1024×1024","price":"$0.09","amount":0.09,"latencyMs":23794,"exampleId":"municipal-dive-supervisor-headshot"}]},{"model":"ideogram-3-0-reframe","air":"ideogram:4@4","name":"Ideogram 3.0 Reframe","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-03-26T00:00:00Z","weight":2,"headline":"Ideogram 3.0 Reframe for style-safe AI outpainting","description":"Ideogram 3.0 Reframe performs style consistent outpainting that extends images beyond their borders. It adapts visuals to new aspect ratios without breaking composition or look. Ideal for repurposing creative, social posts, and design assets for varied formats.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-3-0-reframe.jpg","ogImage":"https://assets.runware.ai/og/ideogram-3-0-reframe.png","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.09,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"expand","price":"$0.06","amount":0.06,"latencyMs":17976,"exampleId":"heritage-bell-foundry-crew-portrait"},{"configuration":"expand","price":"$0.06","amount":0.06,"latencyMs":17394,"exampleId":"sage-beekeeping-suit-colorway-banner"}]},{"model":"ideogram-3-0-replace-background","air":"ideogram:4@5","name":"Ideogram 3.0 Replace Background","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-03-26T00:00:00Z","weight":2,"headline":"AI background replacement for clean product imagery","description":"Ideogram 3.0 Replace Background removes and swaps image backgrounds while preserving key foreground subjects. Ideal for product mockups, marketing assets, and design overlays where consistent scenes and fast iteration matter for teams and creators.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-3-0-replace-background.jpg","ogImage":"https://assets.runware.ai/og/ideogram-3-0-replace-background.png","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.09,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"background-replacement","price":"$0.06","amount":0.06,"latencyMs":18371,"exampleId":"pharmaceutical-robotics-sales-hero"},{"configuration":"background-replacement","price":"$0.06","amount":0.06,"latencyMs":23131,"exampleId":"wildfire-meteorologist-leadership-headshot"},{"configuration":"background-replacement","price":"$0.06","amount":0.06,"latencyMs":31977,"exampleId":"heritage-railway-twilight-campaign-photograph"},{"configuration":"background-replacement","price":"$0.06","amount":0.06,"latencyMs":16008,"exampleId":"winter-glamping-booking-hero"}]},{"model":"rundiffusion-juggernaut-pro-flux","air":"rundiffusion:130@100","name":"Juggernaut Pro Flux","creator":"rundiffusion","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-03-06T00:00:00Z","weight":2,"headline":"Photorealistic Flux based text to image model for pros","description":"Juggernaut Pro Flux is a Flux based text to image model for sharp photorealistic renders. It combines Juggernaut Base with RunDiffusion Photo, improves texture realism, reduces background blur, and preserves depth of field. Built for production grade visual workflows.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/rundiffusion-juggernaut-pro-flux.jpg","ogImage":"https://assets.runware.ai/og/rundiffusion-juggernaut-pro-flux.png","pricingOverview":"~$0.0067 per megapixel","pricingMeasured":[{"configuration":"1024x1024 · 30 steps","price":0.0075},{"configuration":"512x512 · 20 steps","price":0.0042},{"configuration":"512x512 · 28 steps","price":0.0049},{"configuration":"512x512 · 30 steps","price":0.0059},{"configuration":"1024x1024 · 20 steps","price":0.0066},{"configuration":"1024x1024 · 28 steps","price":0.0049},{"configuration":"2048x2048 · 20 steps","price":0.0225},{"configuration":"2048x2048 · 28 steps","price":0.0257},{"configuration":"2048x2048 · 30 steps","price":0.0257},{"configuration":"768x768 · 20 steps","price":0.0059},{"configuration":"768x768 · 28 steps","price":0.0034},{"configuration":"768x768 · 30 steps","price":0.0066},{"configuration":"1280x1280 · 20 steps","price":0.0091},{"configuration":"1280x1280 · 28 steps","price":0.0117},{"configuration":"1280x1280 · 30 steps","price":0.0117},{"configuration":"1536x1536 · 20 steps","price":0.0125},{"configuration":"1536x1536 · 28 steps","price":0.0159},{"configuration":"1536x1536 · 30 steps","price":0.0159}],"pricingExamples":[{"configuration":"image-to-image · 1280×768","price":"$0.00169","amount":0.00169,"latencyMs":3480,"exampleId":"inujima-industrial-heritage-tourism-banner"},{"configuration":"inpaint · 768×1024","price":"$0.00247","amount":0.00247,"latencyMs":5926,"exampleId":"seed-vault-curator-portrait"},{"configuration":"text-to-image · 1024×1024 · steps 33","price":"$0.00247","amount":0.00247,"latencyMs":3533,"exampleId":"cathedral-organ-restoration-clean-plate"},{"configuration":"expand · 720p","price":"$0.00247","amount":0.00247,"latencyMs":5149,"exampleId":"electroacoustic-field-recording-album-cover"}]},{"model":"pixverse-v3-5","air":"pixverse:1@1","name":"PixVerse V3.5","creator":"pixverse","status":"live","releasedAt":"2025-03-01T00:00:00Z","weight":2,"headline":"PixVerse V3.5 early text to video effects model","description":"PixVerse V3.5 provides basic text to video generation with support for visual effects and limited subject motion. It targets short clips for experiments or prototypes. Camera movement is not available, which simplifies control and integration in pipelines.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-v3-5.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v3-5.png","pricingOverview":"Each generation will cost $0.0598/s for 1080p.","pricingRates":[{"amount":0.0224,"unit":"durationSecond","label":"360p"},{"amount":0.0224,"unit":"durationSecond","label":"540p"},{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.0598,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"visual-effect · 540p · 5s","price":"$0.112","amount":0.112,"latencyMs":91270,"exampleId":"anatomical-corsetry-fashion-showcase-reel"},{"configuration":"first-frame · 720p · 8s","price":"$0.299","amount":0.299,"latencyMs":60679,"exampleId":"reflective-cycling-cape-showcase-reel"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":92725,"exampleId":"industrial-techno-foundry-music-segment"},{"configuration":"first-last-frame · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":83977,"exampleId":"countertop-mushroom-grow-kit-reel"}]},{"model":"ideogram-2a-remix","air":"ideogram:2@2","name":"Ideogram 2a Remix","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-02-27T00:00:00Z","weight":2,"headline":"Remix Ideogram 2a images into stylized new variants","description":"Ideogram 2a Remix lets you reinterpret existing images with controlled style shifts. Feed a base reference image. Generate themed variations, visual transformations, and stylized edits while preserving core layout and content. Ideal for brand updates or creative iterations.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-2a-remix.jpg","ogImage":"https://assets.runware.ai/og/ideogram-2a-remix.png","pricingRates":[{"amount":0.025,"unit":"output","label":"Turbo"},{"amount":0.04,"unit":"output","label":"Default"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":5786,"exampleId":"lunar-robotics-field-team-portrait"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":5109,"exampleId":"coastal-memorial-urn-lifestyle-image"}]},{"model":"pixverse-v4","air":"pixverse:1@2","name":"PixVerse V4","creator":"pixverse","status":"live","releasedAt":"2025-02-24T00:00:00Z","weight":2,"headline":"PixVerse V4 AI text to video with pro camera control","description":"PixVerse V4 is a generative video model for text prompts or source images. It improves motion quality and complex camera movement. It adds motion modes, sound effect sync, and style transfer. Ideal for short cinematic clips and rapid creative iteration in production pipelines.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-v4.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v4.png","pricingOverview":"Each generation will cost $0.0598/s for 1080p.","pricingRates":[{"amount":0.0224,"unit":"durationSecond","label":"360p"},{"amount":0.0598,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"visual-effect · 540p · 5s","price":"$0.112","amount":0.112,"latencyMs":80887,"exampleId":"mushroom-farm-open-house-reel"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.299","amount":0.299,"latencyMs":38611,"exampleId":"geothermal-utility-logo-reveal"},{"configuration":"first-frame · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":59732,"exampleId":"mobile-planetarium-social-media-reel"},{"configuration":"text-to-video · 720p · 5s","price":"$0.299","amount":0.299,"latencyMs":49141,"exampleId":"optical-glass-sorting-b-roll"}]},{"model":"open-age-detection","air":"runware:154@1","name":"Open Age Detection","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-02-20T00:00:00Z","weight":2,"headline":"Facial age estimation model","description":"Open Age Detection is a vision model that estimates the age of a person from a facial image. It uses a deep learning classifier trained on diverse face data to predict age categories or age ranges based on visual facial features.","capabilities":["io:image-to-text","op:age-classification","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/open-age-detection.jpg","ogImage":"https://assets.runware.ai/og/open-age-detection.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3406}],"pricingExamples":[{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3731,"exampleId":"senior-esports-division-age-check"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3616,"exampleId":"luxury-campaign-casting-age-classification"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3631,"exampleId":"newspaper-portrait-age-metadata"},{"configuration":"caption","price":"$0.0006","amount":0.0006,"latencyMs":3091,"exampleId":"alcohol-delivery-age-screening"}]},{"model":"minimax-01-director","air":"minimax:2@1","name":"MiniMax 01 Director","creator":"minimax","zdr":true,"status":"live","releasedAt":"2025-01-28T00:00:00Z","weight":2,"headline":"Cinematic text to video with precise camera control","description":"MiniMax 01 Director generates short cinematic video clips from text prompts with director level control. It supports detailed camera movement instructions, stable framing, and reduced motion randomness. Ideal for film previz, ads, and story beats inside production tools.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-01-director.jpg","ogImage":"https://assets.runware.ai/og/minimax-01-director.png","pricingOverview":"Each generation will cost $0.0560/s for 720p.","pricingRates":[{"amount":0.056,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"text-to-video","price":"$0.28","amount":0.28,"latencyMs":164822,"exampleId":"veterinary-blood-bank-symbol-loop"},{"configuration":"text-to-video","price":"$0.28","amount":0.28,"latencyMs":153176,"exampleId":"rooftop-apiary-honey-social-ad"},{"configuration":"first-frame","price":"$0.28","amount":0.28,"latencyMs":166532,"exampleId":"geothermal-control-room-presenter-clip"},{"configuration":"first-frame","price":"$0.28","amount":0.28,"latencyMs":176413,"exampleId":"tidal-workwear-fashion-showcase"}]},{"model":"flex1alpha","air":"runware:160@1","name":"Flex.1-alpha","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-01-18T00:00:00Z","weight":2,"headline":"Compact 8B Apache 2.0 text to image transformer model","description":"Flex.1-alpha is an 8B rectified flow transformer for text to image generation. It matches FLUX.1 style APIs with fewer blocks and lighter compute. It uses an optional guidance embedder so you can run CFG free or keep true CFG for fine tuning in common diffusion toolchains.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/flex1alpha.jpg","ogImage":"https://assets.runware.ai/og/flex1alpha.png"},{"model":"vidu-2-0","air":"vidu:2@0","name":"Vidu 2.0","creator":"vidu","status":"live","releasedAt":"2025-01-15T00:00:00Z","weight":2,"headline":"Fast 1080p AI video generation with strong consistency","description":"Vidu 2.0 is a generative video model for rapid 1080p clip creation. It targets 4 second and 8 second shots with strong subject consistency and support for batch workflows. Developers can drive cinematic clips from text prompts and templates with improved speed and lower cost.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-2-0.jpg","ogImage":"https://assets.runware.ai/og/vidu-2-0.png","pricingOverview":"Each generation will cost $0.0275/s for 720p (4s), or $0.0344/s for 720p (8s).","pricingRates":[{"amount":0.11,"unit":"output","label":"720p · 4s"},{"amount":0.275,"unit":"output","label":"720p · 8s"}],"pricingExamples":[{"configuration":"reference-to-video · 720p · 4s","price":"$0.22","amount":0.22,"latencyMs":60908,"exampleId":"adaptive-handcycle-social-video-ad"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.275","amount":0.275,"latencyMs":127922,"exampleId":"industrial-wool-mill-music-video"},{"configuration":"first-frame · 1080p · 4s","price":"$0.275","amount":0.275,"latencyMs":37105,"exampleId":"ocean-sensor-network-logo-loop"}]},{"model":"bfl-flux-1-depth-dev","air":"runware:103@1","name":"FLUX.1 Depth [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-11-21T00:00:00Z","weight":2,"headline":"Depth guided image generation for precise structure","description":"FLUX.1 Depth [dev] uses input depth maps as structural guidance for new image generation. It preserves scene geometry and layout while allowing creative control over style and details. Ideal for developers who need stable composition in controlled image synthesis.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-depth-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-depth-dev.png"},{"model":"bria-rmbg-v2-0","air":"bria:2@1","name":"Bria RMBG v2.0","creator":"bria","status":"live","releasedAt":"2024-11-13T00:00:00Z","weight":2,"headline":"High precision AI background removal for images","description":"Bria RMBG v2.0 removes complex backgrounds and produces clean alpha mattes for product photos and portraits. It supports varied subjects and lighting. Ideal for ecommerce pipelines and design tools that need fast accurate subject isolation at scale.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/bria-rmbg-v2-0.jpg","ogImage":"https://assets.runware.ai/og/bria-rmbg-v2-0.png","pricingRates":[{"amount":0.018,"unit":"output"}],"pricingExamples":[{"configuration":"remove-background","price":"$0.018","amount":0.018,"latencyMs":3606,"exampleId":"polar-habitat-relighting-cutout"},{"configuration":"remove-background","price":"$0.018","amount":0.018,"latencyMs":4151,"exampleId":"vintage-marionette-archive-cutout"},{"configuration":"remove-background","price":"$0.018","amount":0.018,"latencyMs":5455,"exampleId":"lion-dance-head-compositing-cutout"},{"configuration":"remove-background","price":"$0.018","amount":0.018,"latencyMs":5274,"exampleId":"salmon-fly-ecommerce-cutout"}]},{"model":"flux-1-dev-ip-adapter-xlabs-v3","air":"runware:56@3","name":"FLUX.1 [dev] IP-Adapter (XLabs) v3","status":"live","releasedAt":"2024-11-01T00:00:00Z","weight":2,"headline":"Later XLabs FLUX.1 [dev] IP-Adapter checkpoint for versioned image-reference workflows","description":"FLUX.1 [dev] IP-Adapter (XLabs) v3 is a later versioned checkpoint in the XLabs FLUX IP-Adapter line. It is best understood as a separately exposed iteration of the same FLUX.1 [dev] image-prompt adapter family for workflows that need explicit checkpoint selection by version.","architecture":"flux-1-dev","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-ip-adapter-xlabs-v3.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-ip-adapter-xlabs-v3.png"},{"model":"bfl-flux-1-canny-dev","air":"runware:104@1","name":"FLUX.1 Canny [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-11-01T00:00:00Z","weight":2,"headline":"Edge guided image generation with FLUX.1 Canny Dev","description":"FLUX.1 Canny [dev] is a 12B parameter rectified flow transformer for image generation. It takes a text prompt and an input image. It extracts canny edges as structural guidance. It then generates new images that follow the original composition while applying the prompt.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-canny-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-canny-dev.png"},{"model":"bfl-flux-1-fill-pro","air":"bfl:1@2","name":"FLUX.1 Fill [pro]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2024-11-01T00:00:00Z","weight":2,"headline":"FLUX.1 Fill Pro for Precision AI Inpainting Control","description":"FLUX.1 Fill Pro provides advanced inpainting and outpainting for real and generated images. Supply an input image, mask, and text prompt. The model fills or extends regions with seamless content that matches context and style. Ideal for edits, layout fixes, and content-aware expansion.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-fill-pro.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-fill-pro.png","pricingRates":[{"amount":0.05,"unit":"output"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.05","amount":0.05,"latencyMs":12193,"exampleId":"polar-institute-director-headshot"},{"configuration":"inpaint · steps 32","price":"$0.05","amount":0.05,"latencyMs":16606,"exampleId":"boro-textile-collection-panorama"},{"configuration":"inpaint","price":"$0.05","amount":0.05,"latencyMs":16467,"exampleId":"hydroelectric-intake-tower-night-relight"},{"configuration":"inpaint · steps 30","price":"$0.05","amount":0.05,"latencyMs":15797,"exampleId":"taiko-ensemble-press-photo-update"}]},{"model":"ideogram-2-0-remix","air":"ideogram:3@2","name":"Ideogram 2.0 Remix","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2024-10-23T00:00:00Z","weight":2,"headline":"Remix Ideogram 2.0 images with new moods and styles","description":"Ideogram 2.0 Remix lets you rework existing images while preserving structure and layout. Change styles or mood, adjust composition, and iterate quickly from a reference image. Ideal for designers who need fast visual variants and style exploration from prior outputs.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-2-0-remix.jpg","ogImage":"https://assets.runware.ai/og/ideogram-2-0-remix.png","pricingRates":[{"amount":0.05,"unit":"output","label":"Turbo"},{"amount":0.08,"unit":"output","label":"Default"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.08","amount":0.08,"latencyMs":10253,"exampleId":"antarctic-station-winter-editorial"},{"configuration":"reference-to-image","price":"$0.08","amount":0.08,"latencyMs":17568,"exampleId":"heritage-telescope-tourism-hero-image"}]},{"model":"flux-1-dev-ip-adapter-xlabs-v2","air":"runware:56@2","name":"FLUX.1 [dev] IP-Adapter (XLabs) v2","status":"live","releasedAt":"2024-10-15T00:00:00Z","weight":2,"headline":"Second-generation XLabs FLUX.1 [dev] IP-Adapter with improved training and stronger image conditioning","description":"FLUX.1 [dev] IP-Adapter (XLabs) v2 is the second public XLabs checkpoint in the FLUX IP-Adapter family. XLabs describes it as a more extensively trained version of the original checkpoint, intended for stronger and more stable image-reference conditioning on FLUX.1 [dev].","architecture":"flux-1-dev","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-ip-adapter-xlabs-v2.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-ip-adapter-xlabs-v2.png"},{"model":"flux-1-dev-ip-adapter-xlabs-v1","air":"runware:56@1","name":"FLUX.1 [dev] IP-Adapter (XLabs) v1","status":"live","releasedAt":"2024-10-01T00:00:00Z","weight":2,"headline":"XLabs IP-Adapter checkpoint for FLUX.1 [dev] image-reference conditioning","description":"FLUX.1 [dev] IP-Adapter (XLabs) v1 is the first public XLabs image-prompt adapter checkpoint for the FLUX.1 [dev] family. It adds reference-image conditioning to FLUX workflows so visual cues from an input image can steer subject appearance, style, and composition alongside the prompt.","architecture":"flux-1-dev","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-ip-adapter-xlabs-v1.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-ip-adapter-xlabs-v1.png"},{"model":"minimax-01","air":"minimax:1@1","name":"MiniMax 01","creator":"minimax","zdr":true,"status":"live","releasedAt":"2024-09-01T00:00:00Z","weight":2,"headline":"MiniMax 01 fast text to video model for short clips","description":"MiniMax 01 is a compact text to video model for short clips. It turns simple prompts into 720p videos with smooth motion and cinematic framing. It targets fast iteration and stable output so developers can prototype interactive video features and creative tools with low latency.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/minimax-01.jpg","ogImage":"https://assets.runware.ai/og/minimax-01.png","pricingRates":[{"amount":0.28,"unit":"output"}],"pricingExamples":[{"configuration":"first-frame","price":"$0.28","amount":0.28,"latencyMs":183586,"exampleId":"competitive-topiary-series-title-teaser"},{"configuration":"text-to-video","price":"$0.28","amount":0.28,"latencyMs":174389,"exampleId":"food-fraud-docuseries-title-sting"},{"configuration":"text-to-video","price":"$0.28","amount":0.28,"latencyMs":441563,"exampleId":"clockwork-techno-music-video-segment"},{"configuration":"reference-to-video","price":"$0.42","amount":0.42,"latencyMs":209416,"exampleId":"couture-apiary-suit-showcase-reel"}]},{"model":"flux-1-dev-controlnet-union-pro-blur","air":"runware:28@1","name":"FLUX.1 [dev] ControlNet Union Pro – Blur","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":2,"headline":"FLUX.1 ControlNet mode for deblur and sharpness-guided image restoration","description":"FLUX.1 ControlNet Union Pro – Blur is the blur-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is designed for restoration-style workflows where a blurred or softened reference should guide sharper reconstruction while staying close to the original scene layout.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-blur.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-blur.png"},{"model":"flux-1-dev-controlnet-union-pro-gray","air":"runware:30@1","name":"FLUX.1 [dev] ControlNet Union Pro – Gray","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":2,"headline":"FLUX.1 ControlNet mode for grayscale-guided colorization and tonal reinterpretation","description":"FLUX.1 ControlNet Union Pro – Gray is the grayscale-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is built for colorization and tonal-restoration workflows where a grayscale input should preserve luminance structure while the model generates new color information.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-gray.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-gray.png"},{"model":"flux-1-dev-controlnet-union-pro-low-quality","air":"runware:31@1","name":"FLUX.1 [dev] ControlNet Union Pro – Low Quality","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":2,"headline":"FLUX.1 ControlNet mode for low-quality to high-quality guided enhancement","description":"FLUX.1 ControlNet Union Pro – Low Quality is the low-quality-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is aimed at enhancement workflows where compressed, degraded, or weak-detail inputs should guide a cleaner and more detailed reconstruction without fully discarding the original structure.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-low-quality.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-low-quality.png"},{"model":"flux-1-dev-controlnet-union-instantx","air":"runware:98@1","name":"FLUX.1 [dev] ControlNet Union (InstantX)","status":"live","releasedAt":"2024-08-20T00:00:00Z","weight":2,"headline":"Unified FLUX.1 ControlNet for multi-conditioning across edges, depth, tile, blur, pose, gray, and low-quality inputs","description":"FLUX.1 ControlNet Union (InstantX) is the original unified ControlNet for FLUX.1-dev from InstantX. It combines multiple conditioning types in a single model, supports multi-control workflows, and is designed for broad structural guidance tasks where one ControlNet checkpoint can handle several control modes instead of loading separate specialized adapters.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-instantx.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-instantx.png"},{"model":"sd3-controlnet-canny-instantx","air":"runware:32@1","name":"SD3 ControlNet Canny (InstantX)","status":"live","releasedAt":"2024-08-01T00:00:00Z","weight":2,"headline":"Stable Diffusion 3 ControlNet for edge-guided composition from Canny maps","description":"SD3 ControlNet Canny (InstantX) is a Canny-conditioned ControlNet for Stable Diffusion 3. It is built for workflows where edge maps should define composition, object contours, and scene layout while SD3 handles the final rendering, style, and text-conditioned generation.","architecture":"sd3","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd3-controlnet-canny-instantx.jpg","ogImage":"https://assets.runware.ai/og/sd3-controlnet-canny-instantx.png"},{"model":"sd3-controlnet-depth-instantx","air":"runware:33@1","name":"SD3 ControlNet Depth (InstantX)","status":"live","releasedAt":"2024-08-01T00:00:00Z","weight":2,"headline":"Stable Diffusion 3 ControlNet for geometry-preserving generation from depth maps","description":"SD3 ControlNet Depth (InstantX) is a depth-conditioned ControlNet for Stable Diffusion 3. It is intended for scenes where perspective, depth ordering, and large-scale spatial structure should stay anchored to a source depth map while prompts define the rendered content.","architecture":"sd3","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd3-controlnet-depth-instantx.jpg","ogImage":"https://assets.runware.ai/og/sd3-controlnet-depth-instantx.png"},{"model":"llama-3-1-8b-prompt-enhancer","air":"runware:llama-3.1-8b@prompt-enhancer","name":"Llama 3.1 8B Prompt Enhancer","hosted":true,"status":"live","releasedAt":"2024-07-23T00:00:00Z","weight":2,"headline":"Instruction-tuned language model optimized for prompt enhancement and rewriting","description":"Llama 3.1 8B Prompt Enhancer is a specialized configuration of the 8B instruction-tuned language model designed to refine, expand, and restructure user prompts for downstream generative systems. It improves clarity, specificity, and structural quality while preserving original intent. The model is well suited for enhancing prompts used in text-to-image, text-to-video, and multimodal generation workflows where stronger guidance leads to higher output fidelity.","capabilities":["io:text-to-text","op:prompt-enhance","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/llama-3-1-8b-prompt-enhancer.jpg","ogImage":"https://assets.runware.ai/og/llama-3-1-8b-prompt-enhancer.png"},{"model":"bytedance-seed-audio-1-0","air":"bytedance:seed-audio@1.0","name":"Seed Audio 1.0","creator":"bytedance","status":"live","releasedAt":"2024-06-04T00:00:00Z","weight":2,"headline":"Versatile speech generation model for expressive TTS, voice conversion, and speech editing","description":"Seed Audio 1.0 is a ByteDance speech generation model built for high-quality, highly controllable audio output. Public technical materials describe the underlying Seed-TTS family as supporting zero-shot text-to-speech, cross-lingual speech generation, emotion control, speaker adaptation, voice conversion, speech editing, and speed editing, making it well suited to narration, dubbing, character voice work, and other expressive speech workflows.","capabilities":["io:text-to-audio","io:image-to-audio","io:audio-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-seed-audio-1-0.jpg","ogImage":"https://assets.runware.ai/og/bytedance-seed-audio-1-0.png","pricingOverview":"Pricing is $0.158 per min of Audio generated","pricingRates":[{"amount":0.00263333,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"reference-to-audio","price":"$0.0266","amount":0.026597,"latencyMs":24489,"exampleId":"wheelchair-rugby-broadcast-victory-stinger"},{"configuration":"text-to-audio","price":"$0.0829","amount":0.08295,"latencyMs":37227,"exampleId":"night-garden-admission-stinger"},{"configuration":"text-to-audio","price":"$0.0848","amount":0.084793,"latencyMs":28097,"exampleId":"substation-maintenance-simulator-ambience"},{"configuration":"reference-to-audio","price":"$0.118","amount":0.118237,"latencyMs":78098,"exampleId":"flooded-photograph-rescue-narration"}]},{"model":"clarity","air":"runware:500@1","name":"Clarity","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-04-30T00:00:00Z","weight":2,"headline":"Creative 4× image upscaler for detail enhancement","description":"Clarity is an image upscaling model focused on boosting perceived detail during 2× to 4× enlargement, keeping structure stable while enhancing texture and small features for cleaner, higher-impact images.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/clarity.jpg","ogImage":"https://assets.runware.ai/og/clarity.png","pricingMeasured":[{"configuration":"upscaleFactor 2 · settings.steps 20","price":0.0051,"latencyMs":14820},{"configuration":"upscaleFactor 2 · settings.steps 40","price":0.007,"latencyMs":18037}],"pricingExamples":[{"configuration":"upscale","price":"$0.0026","amount":0.0026,"latencyMs":11385,"exampleId":"historic-organ-campaign-panorama"},{"configuration":"upscale","price":"$0.0032","amount":0.0032,"latencyMs":13001,"exampleId":"antarctic-radio-archive-photo-restoration"},{"configuration":"upscale","price":"$0.0045","amount":0.0045,"latencyMs":15962,"exampleId":"cave-rescue-team-commemorative-wall-portrait"},{"configuration":"upscale","price":"$0.0064","amount":0.0064,"latencyMs":16304,"exampleId":"falconry-hood-cutout-upscale"}]},{"model":"birefnet-matting","air":"runware:112@9","name":"BiRefNet Matting","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-02-12T00:00:00Z","weight":2,"headline":"Alpha matting and soft-edge segmentation with BiRefNet","description":"BiRefNet Matting extends binary segmentation into continuous alpha prediction, enabling smooth transitions for fine structures such as hair, fur, and semi-transparent regions.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-matting.jpg","ogImage":"https://assets.runware.ai/og/birefnet-matting.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":6132}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5609,"exampleId":"salmon-fly-composite-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5894,"exampleId":"glassblower-conference-speaker-headshot-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5339,"exampleId":"kelp-risograph-exhibit-restyle"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5991,"exampleId":"fringed-pendant-virtual-staging-cutout"}]},{"model":"meta-llava-1-6-mistral-7b","air":"runware:150@2","name":"LLaVA-1.6-Mistral-7B","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-31T00:00:00Z","weight":2,"headline":"Vision-language model for image understanding and captioning","description":"LLaVA-1.6-Mistral-7B is a multimodal vision-language model that processes images alongside text to generate descriptive and reasoning-based responses. It enables image captioning and visual understanding by combining a vision encoder with a Mistral 7B language backbone.","capabilities":["io:image-to-text","op:caption","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/meta-llava-1-6-mistral-7b.jpg","ogImage":"https://assets.runware.ai/og/meta-llava-1-6-mistral-7b.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":5646}],"pricingExamples":[{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":22402,"exampleId":"residential-water-damage-assessment"},{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":32076,"exampleId":"offshore-wind-editorial-caption"},{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":14982,"exampleId":"restaurant-editorial-image-caption"},{"configuration":"caption","price":"$0.0045","amount":0.0045,"latencyMs":18380,"exampleId":"runway-show-accessibility-description"}]},{"model":"topazlabs-proteus-4","air":"topazlabs:proteus@4","name":"Topaz Labs Proteus","creator":"topazlabs","zdr":true,"status":"live","releasedAt":"2024-01-16T00:00:00Z","weight":2,"headline":"General-purpose video enhancement for cleaner detail, stronger face recovery, and more consistent upscaling","description":"Proteus is Topaz Labs' general-purpose video enhancement model for upscaling and restoring a wide range of footage. It improves handling of noise, compression artifacts, faces, interlaced material, and scene transitions while maintaining stronger temporal consistency than earlier Proteus versions. Use it for broad video cleanup and enlargement workflows where the source does not call for a more specialized model.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/topazlabs-proteus-4.jpg","ogImage":"https://assets.runware.ai/og/topazlabs-proteus-4.png","pricingMeasured":[{"configuration":"1920×1080 · fps 30 · 4.8s clip","price":0.05,"latencyMs":45117},{"configuration":"3840×2160 · fps 30 · 4.8s clip","price":0.15,"latencyMs":50320},{"configuration":"1920×1080 · fps 60 · 4.8s clip","price":0.15,"latencyMs":59632}],"pricingExamples":[{"configuration":"upscale","price":"$0.05","amount":0.05,"latencyMs":37396,"exampleId":"panoramic-mural-conservation-footage"},{"configuration":"upscale","price":"$0.05","amount":0.05,"latencyMs":71604,"exampleId":"coastal-kelp-harvest-b-roll"},{"configuration":"upscale","price":"$0.05","amount":0.05,"latencyMs":30524,"exampleId":"seismic-observatory-logo-loop"},{"configuration":"upscale","price":"$0.05","amount":0.05,"latencyMs":91336,"exampleId":"seed-bank-curator-briefing"}],"guides":[{"slug":"upscaling-video","title":"Upscaling and enhancing video with Proteus","description":"How to upscale and enhance video with Topaz Proteus: the request shape, width and height as an envelope, the frame rate floor, and what auto mode repairs on its own.","date":"2026-09-03","url":"https://runware.ai/docs/models/topazlabs-proteus-4/guides/upscaling-video"},{"slug":"enhancement-weights","title":"Tuning the enhancement weights","description":"How to tune Topaz Proteus with settings.enhancement: what the auto, relative, and manual modes do, and how the compression, noise, details, and halo weights change a result.","date":"2026-09-03","url":"https://runware.ai/docs/models/topazlabs-proteus-4/guides/enhancement-weights"},{"slug":"grain-and-texture","title":"Keeping grain and texture in an enhanced clip","description":"How to stop Topaz Proteus from over-smoothing: prenoise before the pass, recoverOriginalDetail to blend the source back in, and grain with grainSize on the way out.","date":"2026-09-03","url":"https://runware.ai/docs/models/topazlabs-proteus-4/guides/grain-and-texture"},{"slug":"interlaced-sources","title":"Enhancing interlaced and tape-sourced footage","description":"How to enhance interlaced footage with Topaz Proteus: matching settings.videoType to the source, when settings.fieldOrder matters, and what either one gets wrong.","date":"2026-09-03","url":"https://runware.ai/docs/models/topazlabs-proteus-4/guides/interlaced-sources"},{"slug":"soft-and-blurred-sources","title":"Rescuing soft and out-of-focus footage","description":"How to rescue soft footage with Topaz Proteus: settings.focusFixLevel for out-of-focus subjects, and the preblur and blur weights for aliasing and sharpness.","date":"2026-09-03","url":"https://runware.ai/docs/models/topazlabs-proteus-4/guides/soft-and-blurred-sources"}]},{"model":"ccsr","air":"runware:501@1","name":"CCSR","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-12-23T00:00:00Z","weight":2,"headline":"Diffusion upscaler for high-quality real-world super-resolution","description":"CCSR is a diffusion-based image super-resolution model designed for real-world low-quality inputs, improving sharpness and restoring detail while keeping artifacts controlled during upscaling.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ccsr.jpg","ogImage":"https://assets.runware.ai/og/ccsr.png","pricingMeasured":[{"configuration":"upscaleFactor 2 · settings.steps 20","price":0.0038,"latencyMs":14483},{"configuration":"upscaleFactor 2 · settings.steps 40","price":0.007,"latencyMs":19945}],"pricingExamples":[{"configuration":"upscale","price":"$0.007","amount":0.007,"latencyMs":17398,"exampleId":"deep-sea-specimen-scratchboard-plate"},{"configuration":"upscale","price":"$0.016","amount":0.016,"latencyMs":31837,"exampleId":"bunraku-puppet-stage-composite"},{"configuration":"upscale","price":"$0.0166","amount":0.0166,"latencyMs":38571,"exampleId":"avalanche-rescue-dog-composite"},{"configuration":"upscale","price":"$0.0173","amount":0.0173,"latencyMs":37681,"exampleId":"heritage-fishing-fly-catalog-cleanup"}]},{"model":"rembg-v1-4","air":"runware:109@1","name":"RemBG v1.4","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-09-15T00:00:00Z","weight":2,"headline":"Efficient foreground extraction for image background removal","description":"RemBG v1.4 is an optimized image background removal model focused on rapid and accurate separation of foreground subjects from background. It uses lightweight segmentation to produce clean alpha masks for visual assets, product photos, and portraits without manual editing.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/rembg-v1-4.jpg","ogImage":"https://assets.runware.ai/og/rembg-v1-4.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5303},{"configuration":"settings.alphaMatting true · settings.alphaMattingBackgroundThreshold 10","price":0.0006,"latencyMs":4323}],"pricingExamples":[{"configuration":"foreground-mask","price":"$0.0006","amount":0.0006,"latencyMs":3775,"exampleId":"airport-wildlife-team-portrait-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5884,"exampleId":"sled-dog-arctic-scene-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6303,"exampleId":"archival-dusting-brush-product-cutout"},{"configuration":"foreground-mask","price":"$0.0013","amount":0.0013,"latencyMs":4754,"exampleId":"rare-book-conservator-press-headshot"}]},{"model":"sdxl-ip-adapter-plus-face","air":"runware:55@3","name":"IP-Adapter SDXL Plus-Face","status":"live","releasedAt":"2023-09-08T00:00:00Z","weight":2,"headline":"Stable Diffusion XL IP-Adapter for face-guided portrait generation with stronger identity cues","description":"IP-Adapter SDXL Plus-Face is the face-focused SDXL variant of the Plus adapter family. It uses cropped face conditioning to make portrait, avatar, and character workflows follow facial references more closely while preserving SDXL-level image quality and text compatibility.","architecture":"sdxl","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sdxl-ip-adapter-plus-face.jpg","ogImage":"https://assets.runware.ai/og/sdxl-ip-adapter-plus-face.png"},{"model":"sdxl-controlnet-depth","air":"runware:3@1","name":"SDXL ControlNet Depth","status":"live","releasedAt":"2023-09-07T00:00:00Z","weight":2,"headline":"SDXL ControlNet for depth-guided scene geometry and perspective control","description":"SDXL ControlNet Depth is a depth-conditioned ControlNet trained on Stable Diffusion XL Base 1.0. It is built for workflows where camera perspective, relative depth, and overall scene geometry should follow a depth map while the prompt controls the final rendering.","architecture":"sdxl","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sdxl-controlnet-depth.jpg","ogImage":"https://assets.runware.ai/og/sdxl-controlnet-depth.png"},{"model":"alibaba-qwen2-5-vl-3b-instruct","air":"runware:152@1","name":"Qwen2.5-VL-3B-Instruct","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-08-24T00:00:00Z","weight":2,"headline":"Instruction-tuned vision-language model for image and text understanding","description":"Qwen2.5-VL-3B-Instruct is a multimodal model that processes images and text together to perform visual reasoning, captioning, question answering, and structured output tasks. It integrates a vision encoder with an instruction-tuned language backbone to support complex visual understanding and interactive multimodal responses.","capabilities":["io:image-to-text","op:caption","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen2-5-vl-3b-instruct.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen2-5-vl-3b-instruct.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":5834}],"pricingExamples":[{"configuration":"caption","price":"$0.0013","amount":0.0013,"latencyMs":5419,"exampleId":"marine-conservation-field-report-caption"},{"configuration":"caption","price":"$0.0019","amount":0.0019,"latencyMs":6981,"exampleId":"fashion-editorial-accessibility-caption"},{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":15536,"exampleId":"construction-safety-audit-caption"},{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":13808,"exampleId":"urban-flood-response-caption"}]},{"model":"sdxl-ip-adapter-base","air":"runware:55@1","name":"IP-Adapter SDXL Base","status":"live","releasedAt":"2023-08-18T00:00:00Z","weight":2,"headline":"Stable Diffusion XL IP-Adapter for image-prompt guidance on the SDXL architecture","description":"IP-Adapter SDXL Base is the original IP-Adapter variant for Stable Diffusion XL 1.0. It adds image-prompt conditioning to the SDXL family so reference images can guide composition, mood, and visual identity alongside text prompts.","architecture":"sdxl","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sdxl-ip-adapter-base.jpg","ogImage":"https://assets.runware.ai/og/sdxl-ip-adapter-base.png"},{"model":"runway-aleph-2-0","air":"runway:aleph@2.0","name":"Runway Aleph 2.0","creator":"runway","status":"live","releasedAt":"2026-06-02T00:00:00Z","weight":1,"headline":"Localized video editing with single-frame guidance, multi-shot consistency, and stronger preservation of the original clip","description":"Runway Aleph 2.0 is Runway's upgraded flagship video editing model for transforming existing footage while keeping the rest of the clip stable. It is built for precise localized edits, single-frame visual guidance, and consistent edits across cuts and scene changes, making it useful for product swaps, packaging updates, background changes, relighting, scene cleanup, and other post-production workflows that need the output to stay close to the source video. It supports clips up to 30 seconds at 1080p.","capabilities":["io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/runway-aleph-2-0.jpg","ogImage":"https://assets.runware.ai/og/runway-aleph-2-0.png","pricingOverview":"Pricing starts at $0.28 per second","pricingRates":[{"amount":0.28,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"multiple-keyframes","price":"$1.70","amount":1.6968,"latencyMs":205830,"exampleId":"edwardian-radiochemistry-laboratory-redress"},{"configuration":"first-last-frame","price":"$1.70","amount":1.6968,"latencyMs":186774,"exampleId":"aurora-concert-virtual-set-film"},{"configuration":"first-frame","price":"$1.70","amount":1.6968,"latencyMs":158382,"exampleId":"risograph-seed-archive-museum-film"},{"configuration":"video-edit","price":"$1.70","amount":1.6968,"latencyMs":174613,"exampleId":"submerged-basilica-crew-cleanup"}],"guides":[{"slug":"editing-video","title":"Editing video","description":"How to make localised edits to existing footage with Runway Aleph 2 that change only the targeted region and leave the rest of the clip untouched.","date":"2026-06-02","url":"https://runware.ai/docs/models/runway-aleph-2-0/guides/editing-video"}]},{"model":"lightricks-ltx-2-3","air":"lightricks:ltx@2.3","name":"LTX-2.3","creator":"lightricks","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-03-05T00:00:00Z","weight":1,"headline":"High-fidelity multimodal video generation with native audio","description":"LTX-2.3 is a multimodal video generation model that produces synchronized video and audio from text or images. It supports text-to-video and image-to-video workflows with native dialogue and ambient sound generation, emphasizing temporal stability, strong motion coherence, and production-ready output quality for professional creative pipelines. It can also extend an existing clip with new footage at its start or end.","architecture":"ltx-2-3","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","io:audio-to-video","op:extend","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/lightricks-ltx-2-3.jpg","ogImage":"https://assets.runware.ai/og/lightricks-ltx-2-3.png","pricingRates":[{"amount":0.04,"unit":"durationSecond","label":"720p"},{"amount":0.08,"unit":"durationSecond","label":"1080p"},{"amount":0.16,"unit":"durationSecond","label":"2K"},{"amount":0.1,"unit":"durationSecond","label":"Extend · 1080p, per second added or used as context"}],"pricingExamples":[{"configuration":"first-frame · 720p · 5s","price":"$0.20","amount":0.2,"latencyMs":40628,"exampleId":"salt-pan-percussion-music-video"},{"configuration":"first-last-frame · 720p · 7s","price":"$0.28","amount":0.28,"latencyMs":45785,"exampleId":"bacterial-cellulose-lab-b-roll"},{"configuration":"audio-driven · 1024×1024 · 8s","price":"$0.64","amount":0.64,"latencyMs":54174,"exampleId":"hands-free-jar-opener-demo"},{"configuration":"video-extend · 5s","price":"$0.80","amount":0.8,"latencyMs":59773,"exampleId":"ocean-liner-engine-room-extension"},{"configuration":"multiple-keyframes · 1344×768 · 10s","price":"$0.80","amount":0.8,"latencyMs":68706,"exampleId":"mushroom-farm-post-punk-music-video"},{"configuration":"text-to-video · 1080p · 6s","price":"$0.96","amount":0.96,"latencyMs":71105,"exampleId":"clockwork-chamber-pop-music-video"}]},{"model":"lightricks-ltx-2-3-fast","air":"lightricks:ltx@2.3-fast","name":"LTX-2.3 Fast","creator":"lightricks","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-03-05T00:00:00Z","weight":1,"headline":"Fast multimodal video generation optimized for rapid iteration","description":"LTX-2.3 Fast is a performance-optimized variant of LTX 2.3 designed for rapid video generation with synchronized audio. It supports text-to-video, image-to-video, and audio-conditioned workflows while prioritizing speed, responsiveness, and cost efficiency for draft, preview, and high-velocity creative production use cases.","architecture":"ltx-2-3","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/lightricks-ltx-2-3-fast.jpg","ogImage":"https://assets.runware.ai/og/lightricks-ltx-2-3-fast.png","pricingRates":[{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.06,"unit":"durationSecond","label":"1080p"},{"amount":0.12,"unit":"durationSecond","label":"2K"}],"pricingExamples":[{"configuration":"first-last-frame · 1280×768 · 8s","price":"$0.48","amount":0.48,"latencyMs":33288,"exampleId":"couture-beekeeping-suit-showcase"},{"configuration":"multiple-keyframes · 1536×896 · 10s","price":"$0.60","amount":0.6,"latencyMs":46423,"exampleId":"rare-book-rescue-presenter-clip"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.60","amount":0.6,"latencyMs":52268,"exampleId":"seaweed-biotech-logo-ident"},{"configuration":"audio-driven · 1024×1024 · 12s","price":"$0.72","amount":0.72,"latencyMs":43178,"exampleId":"oyster-hatchery-biosecurity-presenter-clip"},{"configuration":"first-frame · 1080p · 6s","price":"$0.72","amount":0.72,"latencyMs":53450,"exampleId":"wildfire-atlas-logo-motion-loop"}]},{"model":"alibaba-z-image","air":"runware:z-image@0","name":"Z-Image","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-26T00:00:00Z","weight":1,"headline":"Efficient high-quality image generation foundation model","description":"Z-Image is a powerful open-source image generation model with 6 billion parameters built on a scalable single-stream diffusion transformer architecture. It delivers high visual fidelity, strong prompt adherence, and diverse stylistic output for text-to-image and image-to-image tasks, and serves as the full-capacity foundation for distilled variants like Z-Image-Turbo.","architecture":"z-image","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-z-image.jpg","ogImage":"https://assets.runware.ai/og/alibaba-z-image.png","pricingMeasured":[{"configuration":"512×512 · steps 20","price":0.0013,"latencyMs":6175},{"configuration":"1024×1024 · steps 20","price":0.0045,"latencyMs":16846},{"configuration":"1024×1024 · steps 50","price":0.0102,"latencyMs":20374},{"configuration":"2048×2048 · steps 20","price":0.0205,"latencyMs":41999}],"pricingExamples":[{"configuration":"image-to-image · 1280×768","price":"$0.0032","amount":0.0032,"latencyMs":17338,"exampleId":"mycelium-tannery-editorial-portrait"},{"configuration":"inpaint · 896×1152","price":"$0.0045","amount":0.0045,"latencyMs":19568,"exampleId":"chicken-embryo-cutaway-workbook-illustration"},{"configuration":"text-to-image · 1024×1024","price":"$0.0051","amount":0.0051,"latencyMs":14926,"exampleId":"sea-scooter-terracotta-colorway-render"},{"configuration":"expand · 1536×1024","price":"$0.0064","amount":0.0064,"latencyMs":23589,"exampleId":"steppe-archaeology-book-cover"}]},{"model":"bria-fibo-lite","air":"bria:20@3","name":"Bria FIBO Lite","creator":"bria","status":"live","releasedAt":"2026-01-06T00:00:00Z","weight":1,"headline":"Fast, controllable text-to-image generation model","description":"Bria FIBO Lite is a lightweight, efficient text-to-image model designed for rapid image generation with structured prompt control and minimal inference steps. It uses a two-stage distillation approach to deliver high-quality visuals in fewer steps while preserving prompt adherence and visual coherence.","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-fibo-lite.jpg","ogImage":"https://assets.runware.ai/og/bria-fibo-lite.png","pricingRates":[{"amount":0.036,"unit":"output"}],"pricingExamples":[{"configuration":"image-to-image · 1216×832","price":"$0.036","amount":0.036,"latencyMs":10522,"exampleId":"radio-observatory-music-festival-poster"},{"configuration":"image-to-image · 1216×832","price":"$0.036","amount":0.036,"latencyMs":10050,"exampleId":"night-conservatory-reopening-invitation"},{"configuration":"text-to-image · 832×1216","price":"$0.036","amount":0.036,"latencyMs":8425,"exampleId":"cave-surveyor-editorial-profile-avatar"},{"configuration":"text-to-image · 832×1216","price":"$0.036","amount":0.036,"latencyMs":6800,"exampleId":"mycelium-packaging-campaign-visual"}]},{"model":"alibaba-qwen3-tts-1-7b-customvoice","air":"alibaba:qwen@3-tts-1.7b-customvoice","name":"Qwen3-TTS 1.7B CustomVoice","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2026-01-01T00:00:00Z","weight":1,"headline":"Text-to-speech with preset premium timbres and precise style control","description":"Qwen3-TTS 1.7B CustomVoice is a text-to-speech model from Alibaba that offers nine premium preset timbres across various combinations of gender, age, language, and dialect. It provides precise style control over target voices through user instructions, supports voice cloning from a 3-second sample, and generates speech in 10+ languages with latency as low as 97ms.","capabilities":["io:text-to-audio","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen3-tts-1-7b-customvoice.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen3-tts-1-7b-customvoice.png","pricingRates":[{"amount":0.000015,"unit":"character"}],"pricingExamples":[{"configuration":"text-to-audio","price":"$0.0015","amount":0.0015,"latencyMs":2435,"exampleId":"roller-derby-broadcast-vocal-stinger"},{"configuration":"text-to-audio","price":"$0.0048","amount":0.0048,"latencyMs":4563,"exampleId":"geothermal-control-room-voice-bed"},{"configuration":"text-to-audio","price":"$0.0084","amount":0.0084,"latencyMs":7922,"exampleId":"natural-return-service-explainer-voiceover"},{"configuration":"text-to-audio","price":"$0.0106","amount":0.0106,"latencyMs":8149,"exampleId":"seaweed-feed-technology-explainer-voiceover"}]},{"model":"bria-video-eraser","air":"bria:60@1","name":"Bria Video Eraser","creator":"bria","status":"live","releasedAt":"2025-12-19T00:00:00Z","weight":1,"headline":"Motion-aware object removal for video without reshoots","description":"Bria Video Eraser is a video editing model that removes objects from existing video using point-based selection, text instructions, or uploaded masks. It is designed to maintain temporal consistency across frames, prevent flickering and drift, and preserve the original audio track while modifying only the targeted visual regions.","capabilities":["io:video-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-video-eraser.jpg","ogImage":"https://assets.runware.ai/og/bria-video-eraser.png","pricingOverview":"Each second costs $0.05 at 720p.","pricingRates":[{"amount":0.05,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"point-guided-removal","price":"$0.235","amount":0.2354,"latencyMs":57070,"exampleId":"seed-library-ident-cleanup-loop"},{"configuration":"prompted-object-removal","price":"$0.235","amount":0.2354,"latencyMs":58316,"exampleId":"desert-cumbia-lyric-clean-plate"},{"configuration":"automatic-object-removal","price":"$0.235","amount":0.2354,"latencyMs":61829,"exampleId":"mineral-recovery-logo-clean-plate-loop"}]},{"model":"prunaai-p-image","air":"prunaai:1@1","name":"P-Image","creator":"prunaai","status":"live","releasedAt":"2025-12-02T00:00:00Z","weight":1,"headline":"Real-time text-to-image model for production graphics","description":"P-Image is a real-time text-to-image model from Pruna. It delivers sub-second image generation with strong text rendering and tight prompt adherence. It targets production workloads that need fast inference, predictable output control, and efficient scaling through simple API integration.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-image.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-image.png","pricingRates":[{"amount":0.0044,"unit":"output","label":"1024x1024"},{"amount":0.005,"unit":"output","label":"1024x1024 + LoRA"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.0044","amount":0.0044,"latencyMs":4284,"exampleId":"waterlogged-ship-timber-conservation-infographic"},{"configuration":"text-to-image","price":"$0.0044","amount":0.0044,"latencyMs":4716,"exampleId":"counterfeit-saffron-documentary-thumbnail"},{"configuration":"text-to-image","price":"$0.0044","amount":0.0044,"latencyMs":3415,"exampleId":"cold-chain-failure-blog-hero"},{"configuration":"text-to-image","price":"$0.0044","amount":0.0044,"latencyMs":5294,"exampleId":"pigeon-racing-docuseries-key-art"}]},{"model":"klingai-video-o1-pro","air":"klingai:kling@o1","name":"Kling VIDEO O1 Pro","creator":"klingai","status":"live","releasedAt":"2025-12-01T00:00:00Z","weight":1,"headline":"Unified multimodal video model for precise AI editing","description":"Kling VIDEO O1 Pro is a unified multimodal video foundation model for controllable generation and instruction based editing. It supports text prompts, visual references, and video input so developers can build high control pipelines for pacing, transitions, object changes, and style revisions.","capabilities":["io:text-to-video","io:image-to-video","io:video-to-video","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-video-o1-pro.jpg","ogImage":"https://assets.runware.ai/og/klingai-video-o1-pro.png","pricingRates":[{"amount":0.112,"unit":"durationSecond","label":"1080p · T2V"},{"amount":0.112,"unit":"durationSecond","label":"1080p · I2V"},{"amount":0.168,"unit":"durationSecond","label":"1080p · R2V"}],"pricingExamples":[{"configuration":"first-frame · 5s","price":"$0.56","amount":0.56,"latencyMs":121281,"exampleId":"hydropower-spring-thaw-visualization"},{"configuration":"video-edit","price":"$1.01","amount":1.008,"latencyMs":181753,"exampleId":"historic-cinema-facade-cleanup"},{"configuration":"text-to-video · 1080p · 10s","price":"$1.12","amount":1.12,"latencyMs":146875,"exampleId":"competitive-shearer-arena-composite"},{"configuration":"first-last-frame · 10s","price":"$1.12","amount":1.12,"latencyMs":163650,"exampleId":"ocularist-conference-headshot-retouch"},{"configuration":"reference-to-video · 1080p · 7s","price":"$1.18","amount":1.176,"latencyMs":359291,"exampleId":"desert-observatory-nightfall-visualization"}]},{"model":"prunaai-p-image-edit","air":"prunaai:2@1","name":"P-Image-Edit","creator":"prunaai","status":"live","releasedAt":"2025-11-10T00:00:00Z","weight":1,"headline":"High precision multi image AI editor for fast workflows","description":"P-Image-Edit is a real-time image editing model from Pruna AI. It supports multi image refinement, layout control, and style safe transformations while following prompts with high accuracy. Ideal for production pipelines that need consistent edits and tight latency budgets.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/prunaai-p-image-edit.jpg","ogImage":"https://assets.runware.ai/og/prunaai-p-image-edit.png","pricingRates":[{"amount":0.0088,"unit":"output","label":"1024x1024"},{"amount":0.01,"unit":"output","label":"1024x1024 + LoRA"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0088","amount":0.0088,"latencyMs":6744,"exampleId":"dinosaur-exhibition-photo-cleanup"},{"configuration":"reference-to-image","price":"$0.0088","amount":0.0088,"latencyMs":5394,"exampleId":"airport-falconry-campaign-photograph"},{"configuration":"reference-to-image","price":"$0.0088","amount":0.0088,"latencyMs":6956,"exampleId":"space-capsule-catalog-image-cleanup"},{"configuration":"reference-to-image","price":"$0.0088","amount":0.0088,"latencyMs":7002,"exampleId":"arborist-safetywear-virtual-try-on"}]},{"model":"alibaba-wan2-5-preview","air":"runware:201@1","name":"Wan2.5-Preview","creator":"alibaba","status":"live","releasedAt":"2025-09-28T00:00:00Z","weight":1,"headline":"Wan2.5-Preview AI Text to Video with Native Audio","description":"Wan2.5-Preview is Alibaba’s multimodal video model in research preview. It supports text to video and image to video with native audio generation for clips around 10 seconds. It offers strong prompt adherence, smooth motion, and multilingual audio for narrative scenes.","capabilities":["io:text-to-video","io:image-to-video","io:audio-to-video","form:checkpoint","feat:audio"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-5-preview.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-5-preview.png","pricingOverview":"Each generation will cost $0.0946/s for 720p, or $0.1476/s for 1080p.","pricingRates":[{"amount":0.0946,"unit":"durationSecond","label":"720p"},{"amount":0.1476,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.454","amount":0.4538,"latencyMs":108641,"exampleId":"harbor-pilot-safety-bulletin"},{"configuration":"text-to-video · 720p · 5s · audio","price":"$0.454","amount":0.4538,"latencyMs":109488,"exampleId":"florist-shears-social-video-ad"},{"configuration":"audio-driven · 832×1088 · 5s","price":"$0.454","amount":0.4538,"latencyMs":127748,"exampleId":"couture-apiary-protective-suit-reel"},{"configuration":"first-frame · 1080p · 10s · audio","price":"$1.36","amount":1.3573,"latencyMs":188639,"exampleId":"fossil-preparation-tour-reel"}]},{"model":"vidu-q2-pro","air":"vidu:3@1","name":"Vidu Q2 Pro","creator":"vidu","status":"live","releasedAt":"2025-09-25T00:00:00Z","weight":1,"headline":"High fidelity Vidu Q2 Pro model for cinematic AI video","description":"Vidu Q2 Pro is a high fidelity video generation model for cinematic storytelling. It supports text prompts, image inputs, and multi reference control for long form scenes. It targets developers who need controllable motion, stable characters, and smooth camera work for complex shots.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/vidu-q2-pro.jpg","ogImage":"https://assets.runware.ai/og/vidu-q2-pro.png","pricingOverview":"Each generation will cost $0.0688/s for 1080p.","pricingRates":[{"amount":0.09625,"unit":"output","label":"720p · 5s"},{"amount":0.1375,"unit":"output","label":"720p · 8s"},{"amount":0.165,"unit":"output","label":"1080p · 5s"},{"amount":0.2475,"unit":"output","label":"1080p · 8s"}],"pricingExamples":[{"configuration":"reference-to-video · 960×960 · 5s","price":"$0.124","amount":0.12375,"latencyMs":59564,"exampleId":"coral-micropropagation-laboratory-b-roll"},{"configuration":"first-frame · 720p · 6s","price":"$0.179","amount":0.17875,"latencyMs":91183,"exampleId":"flower-market-music-video-segment"},{"configuration":"first-last-frame · 720p · 8s","price":"$0.234","amount":0.23375,"latencyMs":236266,"exampleId":"mine-safety-avatar-presenter-clip"}]},{"model":"alibaba-wan2-5-preview-image","air":"runware:201@10","name":"Wan2.5-Preview Image","creator":"alibaba","status":"live","releasedAt":"2025-09-24T00:00:00Z","weight":1,"headline":"High fidelity Wan2.5 image generation for rich single frames","description":"Wan2.5-Preview Image is a single frame generator built from the Wan2.5 video stack. It focuses on detailed depth structure, strong prompt following, multilingual text rendering, and video grade visual quality for production ready stills in creative or product workflows.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-wan2-5-preview-image.jpg","ogImage":"https://assets.runware.ai/og/alibaba-wan2-5-preview-image.png","pricingOverview":"$0.027 per image","pricingRates":[{"amount":0.027,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.03","amount":0.03,"latencyMs":33613,"exampleId":"sewer-robotics-magazine-cover"},{"configuration":"text-to-image","price":"$0.03","amount":0.03,"latencyMs":16627,"exampleId":"planetarium-curator-profile-avatar"},{"configuration":"text-to-image","price":"$0.03","amount":0.03,"latencyMs":17238,"exampleId":"seismic-ceramics-exhibition-book-cover"},{"configuration":"text-to-image","price":"$0.03","amount":0.03,"latencyMs":33346,"exampleId":"independent-perfumer-editorial-portrait"}]},{"model":"alibaba-qwen-image-edit-plus","air":"runware:108@22","name":"Qwen-Image-Edit-Plus","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-09-23T00:00:00Z","weight":1,"headline":"Advanced multi image editing with precise ControlNet guidance","description":"Qwen-Image-Edit-Plus is a 20B image editing model that supports multi image workflows and strong identity preservation. It improves consistency on single image edits and adds native ControlNet style conditioning for precise structure control, layout edits, and bilingual text manipulation.","architecture":"qwen-image-edit-plus","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-edit-plus.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-edit-plus.png","pricingMeasured":[{"configuration":"1024x1024 · 30 steps","price":0.0166},{"configuration":"1K","price":0.03}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":19025,"exampleId":"carousel-horse-catalog-cleanup"},{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":11694,"exampleId":"antarctic-logistics-winter-report-hero"},{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":11886,"exampleId":"mobile-apiary-brochure-hero"},{"configuration":"reference-to-image","price":"$0.0141","amount":0.0141,"latencyMs":12287,"exampleId":"solar-farm-raptor-report-hero"}]},{"model":"bria-image-replace-background","air":"bria:11@1","name":"Bria Image Replace Background","creator":"bria","status":"live","releasedAt":"2025-09-18T00:00:00Z","weight":1,"headline":"AI background replacement for production ready images","description":"Bria Image Replace Background replaces image backgrounds from text prompts or reference images. It preserves the main subject with high fidelity. Ideal for automated product visuals, creative workflows, and bulk content pipelines.","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-image-replace-background.jpg","ogImage":"https://assets.runware.ai/og/bria-image-replace-background.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"background-replacement","price":"$0.04","amount":0.04,"latencyMs":18893,"exampleId":"linocut-wetland-scientist-portrait"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":10487,"exampleId":"coral-avalanche-transceiver-colorway-catalog"},{"configuration":"background-replacement","price":"$0.04","amount":0.04,"latencyMs":16018,"exampleId":"polar-night-sauna-campaign-image"},{"configuration":"reference-to-image","price":"$0.04","amount":0.04,"latencyMs":11691,"exampleId":"arctic-radio-operator-portrait-restoration"}]},{"model":"flux-1-dev-srpo","air":"runware:111@1","name":"FLUX.1 [dev] SRPO","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-09-12T00:00:00Z","weight":1,"headline":"FLUX.1 SRPO dev for high realism text to image","description":"FLUX.1 [dev] SRPO is a 12B flow transformer finetuned with Tencent SRPO for higher realism and aesthetics in text guided image generation. It improves lighting, texture, and artifact control. Ideal for teams that need controllable, high quality image output from text prompts.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-srpo.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-srpo.png","pricingMeasured":[{"configuration":"512×512 · steps 20","price":0.0006,"latencyMs":1713},{"configuration":"1024×1024 · steps 40","price":0.0019,"latencyMs":3963}],"pricingExamples":[{"configuration":"text-to-image · 720p · steps 28","price":"$0.0013","amount":0.0013,"latencyMs":2435,"exampleId":"marine-induction-cooktop-lifestyle-hero"},{"configuration":"image-to-image · 1024×1024 · steps 26","price":"$0.0013","amount":0.0013,"latencyMs":4353,"exampleId":"urban-heat-mitigation-explainer"},{"configuration":"reference-to-image · 1024×1024 · steps 28","price":"$0.0019","amount":0.0019,"latencyMs":4595,"exampleId":"tardigrade-dormancy-science-activity-card"},{"configuration":"inpaint · 1152×896 · steps 30","price":"$0.0019","amount":0.0019,"latencyMs":5600,"exampleId":"audiology-clinic-reception-renovation"},{"configuration":"expand · 1216×832 · steps 28","price":"$0.0032","amount":0.0032,"latencyMs":20216,"exampleId":"leafcutter-ant-colony-reader-spread"}]},{"model":"bria-video-increase-resolution","air":"bria:50@1","name":"Bria Video Increase Resolution","creator":"bria","status":"live","releasedAt":"2025-09-11T00:00:00Z","weight":1,"headline":"Bria video upscaling for sharper high resolution output","description":"Bria Video Increase Resolution upscales visual assets by 2x or 4x while preserving detail and transparency. Ideal for product footage and design workflows that need clean high resolution results without manual rework or artifacts.","capabilities":["io:video-to-video","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-video-increase-resolution.jpg","ogImage":"https://assets.runware.ai/og/bria-video-increase-resolution.png","pricingOverview":"Each operation costs $0.05/s.","pricingRates":[{"amount":0.05,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"upscale","price":"$0.302","amount":0.3021,"latencyMs":55909,"exampleId":"luxury-bookbinding-process-reel"},{"configuration":"upscale","price":"$0.302","amount":0.3021,"latencyMs":56825,"exampleId":"arctic-seed-vault-documentary-teaser"},{"configuration":"upscale","price":"$0.302","amount":0.3021,"latencyMs":55077,"exampleId":"illuminated-manuscript-conservation-master"},{"configuration":"upscale","price":"$0.302","amount":0.3021,"latencyMs":57639,"exampleId":"sculptural-chainmail-handbag-showcase-reel"}]},{"model":"pixverse-v5","air":"pixverse:1@5","name":"PixVerse V5","creator":"pixverse","status":"live","releasedAt":"2025-08-29T00:00:00Z","weight":1,"headline":"PixVerse V5 cinematic text to video and image to video","description":"PixVerse V5 generates high fidelity video from text prompts or single images. It delivers smooth motion and sharp cinematic frames with strong prompt alignment. Ideal for creators who need fast iteration, keyframe control, and consistent style across shots.","capabilities":["io:text-to-video","io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/pixverse-v5.jpg","ogImage":"https://assets.runware.ai/og/pixverse-v5.png","pricingOverview":"Each generation will cost $0.0598/s for 1080p.","pricingRates":[{"amount":0.0224,"unit":"durationSecond","label":"360p"},{"amount":0.0224,"unit":"durationSecond","label":"540p"},{"amount":0.03,"unit":"durationSecond","label":"720p"},{"amount":0.0598,"unit":"durationSecond","label":"1080p"}],"pricingExamples":[{"configuration":"first-last-frame · 540p · 5s","price":"$0.112","amount":0.112,"latencyMs":47186,"exampleId":"precision-fermentation-sampling-b-roll"},{"configuration":"first-frame · 720p · 8s","price":"$0.299","amount":0.299,"latencyMs":64982,"exampleId":"nocturnal-transit-logo-motion-loop"},{"configuration":"visual-effect · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":49260,"exampleId":"couture-beekeeping-suit-rotation-reel"},{"configuration":"text-to-video · 1080p · 5s","price":"$0.299","amount":0.299,"latencyMs":101719,"exampleId":"backwater-valve-cutaway-explainer"}]},{"model":"alibaba-qwen-image-edit","air":"runware:108@20","name":"Qwen-Image-Edit","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-08-19T00:00:00Z","weight":1,"headline":"High fidelity text guided image editing for Qwen","description":"Qwen-Image-Edit is an instruction based image editing model built on the 20B Qwen-Image foundation. It performs semantic edits and local appearance changes while preserving layout and text fidelity. Ideal for programmatic asset cleanup, style tweaks, and precise bilingual text updates.","architecture":"qwen-image-edit","capabilities":["io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen-image-edit.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen-image-edit.png","pricingOverview":"Each image generation costs $0.0032 at 1024x1024.","pricingRates":[{"amount":0.0032,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.0045","amount":0.0045,"latencyMs":8988,"exampleId":"polar-scientist-speaker-headshot"},{"configuration":"reference-to-image","price":"$0.0045","amount":0.0045,"latencyMs":8974,"exampleId":"oyster-farmer-wader-virtual-try-on"},{"configuration":"reference-to-image","price":"$0.0045","amount":0.0045,"latencyMs":9556,"exampleId":"lighthouse-lens-workshop-photo-restoration"},{"configuration":"reference-to-image","price":"$0.0045","amount":0.0045,"latencyMs":9471,"exampleId":"puppet-theater-cast-press-portrait"}]},{"model":"runway-gen-4-image-turbo","air":"runway:4@2","name":"Runway Gen-4 Image Turbo","creator":"runway","status":"live","releasedAt":"2025-08-19T00:00:00Z","weight":1,"headline":"Fast Gen-4 image generation for rapid creative iteration","description":"Runway Gen-4 Image Turbo is a faster Gen-4 image model for teams that need quick visual iteration. Generate concepts in seconds from text prompts or references while preserving key style and composition control. Ideal for testing ideas before higher cost image workflows.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/runway-gen-4-image-turbo.jpg","ogImage":"https://assets.runware.ai/og/runway-gen-4-image-turbo.png","pricingRates":[{"amount":0.02,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image","price":"$0.02","amount":0.02,"latencyMs":24684,"exampleId":"winter-waterworks-relighting-visualization"},{"configuration":"reference-to-image","price":"$0.02","amount":0.02,"latencyMs":21340,"exampleId":"historic-opera-house-cleanup"},{"configuration":"reference-to-image","price":"$0.02","amount":0.02,"latencyMs":25769,"exampleId":"marbled-paper-atelier-editorial-panorama"},{"configuration":"reference-to-image","price":"$0.02","amount":0.02,"latencyMs":19703,"exampleId":"sound-archive-exhibition-hero-image"}]},{"model":"krea-flux-1-krea-dev","air":"runware:107@1","name":"FLUX.1 Krea [dev]","creator":"krea","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-07-31T00:00:00Z","weight":1,"headline":"FLUX.1 Krea Dev for photorealistic open-weight generation","description":"FLUX.1 Krea [dev] is an open-weight text-to-image model from Black Forest Labs and Krea AI. It targets opinionated aesthetics and realistic photography. Developers can drop it into FLUX.1 dev workflows to build custom generators that avoid the typical AI look.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-krea-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-krea-dev.png","pricingOverview":"Each image generation costs $0.0098 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 28 steps","price":0.0098}],"pricingExamples":[{"configuration":"control-guided · 1152×768","price":"$0.0073","amount":0.0073,"latencyMs":21780,"exampleId":"maritime-folk-cassette-cover"},{"configuration":"image-to-image · 1152×768","price":"$0.0073","amount":0.0073,"latencyMs":15477,"exampleId":"oyster-farm-salinity-meter-campaign"},{"configuration":"reference-to-image · 1024×1024","price":"$0.0073","amount":0.0073,"latencyMs":15496,"exampleId":"cave-rescue-wayfinding-icon-set"},{"configuration":"text-to-image · 1024×1024 · steps 28","price":"$0.0079","amount":0.0079,"latencyMs":3697,"exampleId":"museum-ceramic-conservator-avatar"},{"configuration":"expand · 720p","price":"$0.0079","amount":0.0079,"latencyMs":4470,"exampleId":"airport-falconry-explainer-thumbnail"},{"configuration":"inpaint · 1024×1024","price":"$0.0079","amount":0.0079,"latencyMs":5669,"exampleId":"agricultural-bio-thriller-series-poster"}]},{"model":"bfl-flux-1-kontext-max","air":"bfl:4@1","name":"FLUX.1 Kontext [max]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2025-07-11T00:00:00Z","weight":1,"headline":"High fidelity FLUX.1 Kontext max for precise image edits","description":"FLUX.1 Kontext [max] is a high quality text to image model for production workflows. It focuses on prompt accuracy, sharp local edits, and premium typography rendering. Use it for detailed visual design, branded visuals, and consistent character safe image generation.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-kontext-max.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-kontext-max.png","pricingOverview":"Each image generation costs $0.08 at 1024x1024.","pricingRates":[{"amount":0.08,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 880×1184","price":"$0.08","amount":0.08,"latencyMs":17025,"exampleId":"alpine-observatory-winter-campaign-hero"},{"configuration":"text-to-image · 1392×752","price":"$0.08","amount":0.08,"latencyMs":9051,"exampleId":"rooftop-beekeeping-suit-try-on"},{"configuration":"reference-to-image · 880×1184","price":"$0.08","amount":0.08,"latencyMs":12387,"exampleId":"maritime-auction-catalog-portrait"},{"configuration":"text-to-image · 1392×752","price":"$0.08","amount":0.08,"latencyMs":12584,"exampleId":"opera-costume-atelier-panorama"}]},{"model":"bfl-flux-1-kontext-pro","air":"bfl:3@1","name":"FLUX.1 Kontext [pro]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2025-07-11T00:00:00Z","weight":1,"headline":"Context aware FLUX.1 image editing and generation model","description":"FLUX.1 Kontext [pro] combines fast text to image generation with precise image editing. It supports reference images, local region edits, and full scene changes while preserving style and character identity. Ideal for iterative workflows in design, product visuals, and storytelling pipelines.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-kontext-pro.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-kontext-pro.png","pricingOverview":"Each image generation costs $0.04 at 1024x1024.","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"reference-to-image · 880×1184","price":"$0.04","amount":0.04,"latencyMs":8048,"exampleId":"air-ambulance-incubator-deployment-visual"},{"configuration":"text-to-image · 1392×752","price":"$0.04","amount":0.04,"latencyMs":10726,"exampleId":"aquaculture-sensor-campaign-hero"},{"configuration":"reference-to-image · 880×1184","price":"$0.04","amount":0.04,"latencyMs":8650,"exampleId":"tattoo-machine-cobalt-colorway"},{"configuration":"text-to-image · 1392×752","price":"$0.04","amount":0.04,"latencyMs":7487,"exampleId":"avalanche-patrol-team-composite"}]},{"model":"bria-image-increase-resolution","air":"bria:52@1","name":"Bria Image Increase Resolution","creator":"bria","status":"live","releasedAt":"2025-07-08T00:00:00Z","weight":1,"headline":"High quality 2x and 4x AI image upscaling by Bria","description":"Bria Image Increase Resolution is a specialized upscaling model for production workflows. It increases image resolution by 2x or 4x while preserving fine detail and transparency. Use it to upscale illustrations, product shots or UI assets without artifacts.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-image-increase-resolution.jpg","ogImage":"https://assets.runware.ai/og/bria-image-increase-resolution.png","pricingOverview":"Each operation costs $0.018.","pricingRates":[{"amount":0.018,"unit":"output"}],"pricingExamples":[{"configuration":"upscale","price":"$0.04","amount":0.04,"latencyMs":8843,"exampleId":"utility-lineworker-recruitment-portrait"},{"configuration":"upscale","price":"$0.04","amount":0.04,"latencyMs":8606,"exampleId":"archival-cephalopod-plate-enlargement"},{"configuration":"upscale","price":"$0.04","amount":0.04,"latencyMs":10333,"exampleId":"cassowary-museum-campaign-composite"},{"configuration":"upscale","price":"$0.04","amount":0.04,"latencyMs":7168,"exampleId":"blue-hour-observatory-hero-image"}]},{"model":"bfl-flux-1-kontext-dev","air":"runware:106@1","name":"FLUX.1 Kontext [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-06-17T00:00:00Z","weight":1,"headline":"Open image editing model for fast iterative workflows","description":"FLUX.1 Kontext [dev] is an open weights image editing model tuned for fast iterative development. It supports local edits and full scene changes from prompts. Use it for style transfer, background swaps, object edits, and character consistent variations in your pipelines.","architecture":"flux-1-kontext-dev","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-kontext-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-kontext-dev.png","pricingOverview":"Each image generation costs $0.0105 at 1024x1024.","pricingMeasured":[{"configuration":"1024×1024 · steps 28","price":0.0098,"latencyMs":27298},{"configuration":"1024×1024 · steps 50","price":0.0124,"latencyMs":11594}],"pricingExamples":[{"configuration":"text-to-image · 1248×832","price":"$0.0098","amount":0.0098,"latencyMs":6633,"exampleId":"coastal-floodgate-winter-lighting-visualization"},{"configuration":"text-to-image · 1248×832","price":"$0.0098","amount":0.0098,"latencyMs":6996,"exampleId":"subsea-robot-reef-survey-composite"},{"configuration":"reference-to-image · 832×1248","price":"$0.013","amount":0.013,"latencyMs":37767,"exampleId":"foley-artist-editorial-headshot"},{"configuration":"reference-to-image · 832×1248","price":"$0.013","amount":0.013,"latencyMs":13224,"exampleId":"urban-raptor-conservation-portrait"}]},{"model":"bria-video-background-removal","air":"bria:51@1","name":"Bria Video Background Removal","creator":"bria","status":"live","releasedAt":"2025-06-14T00:00:00Z","weight":1,"headline":"Real time AI video background removal for clean layers","description":"Bria Video Background Removal detects moving subjects in video and removes complex backgrounds in real time. It outputs clean alpha channels for precise keying. Ideal for virtual production, live streaming, post production pipelines, and automated video workflows.","capabilities":["io:video-to-video","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/bria-video-background-removal.jpg","ogImage":"https://assets.runware.ai/og/bria-video-background-removal.png","pricingOverview":"Each operation costs $0.05/s.","pricingRates":[{"amount":0.05,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"remove-background","price":"$0.302","amount":0.3024,"latencyMs":23143,"exampleId":"adaptive-fencing-reel-subject-cutout"},{"configuration":"remove-background","price":"$0.302","amount":0.3024,"latencyMs":21844,"exampleId":"punk-asl-lyric-cutout"},{"configuration":"remove-background","price":"$0.302","amount":0.3024,"latencyMs":23159,"exampleId":"tidal-turbine-logo-loop"},{"configuration":"remove-background","price":"$0.302","amount":0.3024,"latencyMs":37057,"exampleId":"polar-docuseries-title-sequence-plate"}]},{"model":"bria-3-2","air":"bria:10@1","name":"Bria 3.2","creator":"bria","status":"live","releasedAt":"2025-06-10T00:00:00Z","weight":1,"headline":"Commercial-safe text to image model for production use","description":"Bria 3.2 is a compact text to image model built on fully licensed data. It delivers strong prompt alignment, high aesthetic quality, and reliable short text rendering. Ideal for enterprise workflows that need compliant image generation with predictable behavior and easy integration.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bria-3-2.jpg","ogImage":"https://assets.runware.ai/og/bria-3-2.png","pricingOverview":"Each image generation costs $0.04 at 1024x1024.","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":12649,"exampleId":"soil-dna-kit-packaging"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":16155,"exampleId":"critical-minerals-lab-editorial-portrait"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":8069,"exampleId":"pediatric-mri-ceiling-mural"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":13576,"exampleId":"falconry-gps-tracker-packshot"}]},{"model":"runway-gen-4-image","air":"runway:4@1","name":"Runway Gen-4 Image","creator":"runway","status":"live","releasedAt":"2025-05-07T00:00:00Z","weight":1,"headline":"High-fidelity Gen-4 image generation with style control","description":"Runway Gen-4 Image is a text-to-image model for production work. It offers strong prompt adherence, fine stylistic control, and visual consistency across scenes and characters. Ideal for pipelines that link still images into video while preserving look and layout.","capabilities":["io:text-to-image","io:image-to-image","op:edit","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/runway-gen-4-image.jpg","ogImage":"https://assets.runware.ai/og/runway-gen-4-image.png","pricingRates":[{"amount":0.05,"unit":"output","label":"720x720 · 720p"},{"amount":0.08,"unit":"output","label":"1024x1024 · 1K"},{"amount":0.08,"unit":"output","label":"1080x1080 · 1080p"}],"pricingExamples":[{"configuration":"reference-to-image · 1080p","price":"$0.08","amount":0.08,"latencyMs":37656,"exampleId":"maritime-artifact-exhibition-hero"},{"configuration":"reference-to-image · 1080p","price":"$0.08","amount":0.08,"latencyMs":71118,"exampleId":"wildfire-strategy-game-key-art"},{"configuration":"text-to-image · 1360×768","price":"$0.08","amount":0.08,"latencyMs":29571,"exampleId":"opera-wardrobe-banner-expansion"},{"configuration":"text-to-image · 1360×768","price":"$0.08","amount":0.08,"latencyMs":28961,"exampleId":"urban-beekeeper-editorial-headshot"}]},{"model":"ideogram-3-0-edit","air":"ideogram:4@3","name":"Ideogram 3.0 Edit","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-05-01T00:00:00Z","weight":1,"headline":"High precision Ideogram 3.0 image inpainting model","description":"Ideogram 3.0 Edit lets you inpaint images with surgical control. Upload an image, mask a region, then refine layout or text while the rest stays intact. Ideal for typography fixes, layout tweaks, brand updates, and production safe visual polish in existing assets.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-3-0-edit.jpg","ogImage":"https://assets.runware.ai/og/ideogram-3-0-edit.png","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.09,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"inpaint","price":"$0.06","amount":0.06,"latencyMs":16948,"exampleId":"railway-signalwoman-portrait-restoration"},{"configuration":"identity-preservation","price":"$0.06","amount":0.06,"latencyMs":26268,"exampleId":"volcanologist-field-report-portrait"},{"configuration":"masked-identity-preservation","price":"$0.06","amount":0.06,"latencyMs":24318,"exampleId":"synchronized-swim-roster-portrait"}]},{"model":"ideogram-3-0-remix","air":"ideogram:4@2","name":"Ideogram 3.0 Remix","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-04-30T00:00:00Z","weight":1,"headline":"Remix Ideogram 3.0 images with fresh styles and palettes","description":"Ideogram 3.0 Remix lets you rework existing images into new styles while it preserves layout and composition. Use it to test creative variants, adjust palettes, or adapt designs for new campaigns. Ideal for A/B testing, rapid iteration, and brand safe visual updates.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-3-0-remix.jpg","ogImage":"https://assets.runware.ai/og/ideogram-3-0-remix.png","pricingRates":[{"amount":0.03,"unit":"output","label":"Turbo"},{"amount":0.06,"unit":"output","label":"Default"},{"amount":0.09,"unit":"output","label":"Quality"}],"pricingExamples":[{"configuration":"reference-to-image · 1024×1024","price":"$0.06","amount":0.06,"latencyMs":12620,"exampleId":"beekeeping-cooperative-textile-campaign-artwork"},{"configuration":"masked-identity-preservation · 896×1152","price":"$0.06","amount":0.06,"latencyMs":22975,"exampleId":"adaptive-rain-cape-virtual-try-on"},{"configuration":"identity-preservation · 832×1216","price":"$0.09","amount":0.09,"latencyMs":25163,"exampleId":"subsea-engineer-leadership-headshot"}]},{"model":"runway-gen-4-turbo","air":"runway:1@1","name":"Runway Gen-4 Turbo","creator":"runway","status":"live","releasedAt":"2025-04-08T00:00:00Z","weight":1,"headline":"High speed Gen-4 Turbo image to video generation","description":"Runway Gen-4 Turbo is a high speed variant of Gen-4 for rapid video ideation. It turns reference images into short cinematic clips with strong character consistency, smooth motion, and reduced credit cost. Ideal for fast iteration in production and previsualization pipelines.","capabilities":["io:image-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/runway-gen-4-turbo.jpg","ogImage":"https://assets.runware.ai/og/runway-gen-4-turbo.png","pricingOverview":"$0.05 / second of video ","pricingRates":[{"amount":0.05,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"first-frame","price":"$0.252","amount":0.2521,"latencyMs":27531,"exampleId":"urban-pigeon-racing-series-teaser"},{"configuration":"first-frame","price":"$0.252","amount":0.2521,"latencyMs":24645,"exampleId":"pre-dawn-fish-auction-music-video"},{"configuration":"first-frame","price":"$0.252","amount":0.2521,"latencyMs":24919,"exampleId":"seed-cryobank-training-presenter-clip"},{"configuration":"first-frame","price":"$0.252","amount":0.2521,"latencyMs":25325,"exampleId":"pipe-organ-documentary-title-sequence"}]},{"model":"rundiffusion-juggernaut-base-flux","air":"rundiffusion:120@100","name":"Juggernaut Base Flux","creator":"rundiffusion","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-03-06T00:00:00Z","weight":1,"headline":"Flux Dev compatible model for sharper photoreal images","description":"Juggernaut Base Flux is a finetuned Flux Dev compatible model for high fidelity image generation. It improves detail and contrast while keeping LoRA and LoCON workflows intact. Use it as a drop in upgrade in existing pipelines that target Flux Dev style text to image generation.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/rundiffusion-juggernaut-base-flux.jpg","ogImage":"https://assets.runware.ai/og/rundiffusion-juggernaut-base-flux.png","pricingMeasured":[{"configuration":"512×512 · steps 33","price":0.00247,"latencyMs":2883},{"configuration":"1024×1024 · steps 33","price":0.00585,"latencyMs":8725},{"configuration":"1024×1024 · steps 50","price":0.00832,"latencyMs":10624},{"configuration":"2048×2048 · steps 33","price":0.02496,"latencyMs":31661}],"pricingExamples":[{"configuration":"control-guided · 1024×768","price":"$0.00416","amount":0.00416,"latencyMs":24581,"exampleId":"mesopotamian-marshes-tourism-hero"},{"configuration":"image-to-image · 1024×1024","price":"$0.00416","amount":0.00416,"latencyMs":14347,"exampleId":"civic-infrastructure-magazine-cover"},{"configuration":"inpaint · 896×1152","price":"$0.00494","amount":0.00494,"latencyMs":8871,"exampleId":"quarry-hearing-protection-campaign"},{"configuration":"text-to-image · 720p","price":"$0.00494","amount":0.00494,"latencyMs":11827,"exampleId":"oyster-reef-benefit-dinner-invitation"},{"configuration":"expand · 1536×1024","price":"$0.00754","amount":0.00754,"latencyMs":17439,"exampleId":"night-greenhouse-ambient-album-cover"}]},{"model":"ideogram-2a","air":"ideogram:2@1","name":"Ideogram 2a","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2025-02-27T00:00:00Z","weight":1,"headline":"Fast text to image generation for design and photography","description":"Ideogram 2a is a fast text to image model built for layouts that need clear structure and legible text. It improves prompt following, spatial control, and subject placement. Use it for graphic design workflows, product shots, logos, posters, and quick visual iterations through the API.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-2a.jpg","ogImage":"https://assets.runware.ai/og/ideogram-2a.png","pricingRates":[{"amount":0.025,"unit":"output","label":"Turbo"},{"amount":0.04,"unit":"output","label":"Default"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":12513,"exampleId":"polar-field-recording-album-cover"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":12441,"exampleId":"subsea-cable-documentary-thumbnail"}]},{"model":"bytedance-omnihuman-1","air":"bytedance:5@1","name":"OmniHuman-1","creator":"bytedance","status":"live","releasedAt":"2025-02-05T00:00:00Z","weight":1,"headline":"Generate realistic talking human videos from one image","description":"OmniHuman-1 is a ByteDance research model for human video generation from a single image and motion signals like audio. It focuses on accurate lip sync, expressive motion, and strong generalization across portraits, full body shots, cartoons, and stylized avatars.","capabilities":["io:image-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bytedance-omnihuman-1.jpg","ogImage":"https://assets.runware.ai/og/bytedance-omnihuman-1.png","pricingRates":[{"amount":0.0924,"unit":"durationSecond"}],"pricingExamples":[{"configuration":"audio-driven-photo-avatar","price":"$0.418","amount":0.4176,"latencyMs":123213,"exampleId":"mycelium-packaging-mascot-logo-sting"},{"configuration":"audio-driven-photo-avatar","price":"$0.488","amount":0.4879,"latencyMs":344424,"exampleId":"heritage-seed-library-logo-sting"},{"configuration":"audio-driven-photo-avatar","price":"$0.61","amount":0.6098,"latencyMs":140521,"exampleId":"polar-researcher-editorial-stock-clip"},{"configuration":"audio-driven-photo-avatar","price":"$1.09","amount":1.0866,"latencyMs":165673,"exampleId":"wholesale-fish-auction-caller-clip"}]},{"model":"flux-1-fill-dev-onereward","air":"runware:121@1","name":"FLUX.1 Fill [dev] OneReward","hosted":true,"zdr":true,"status":"live","releasedAt":"2025-01-15T00:00:00Z","weight":1,"headline":"RLHF enhanced FLUX.1 Fill for precise image inpainting","description":"FLUX.1 Fill [dev] OneReward is an experimental FLUX.1 Fill variant trained with the OneReward preference model. It targets context aware inpainting, compositional image extension, object removal, and structure aligned repairs that follow learned human aesthetic preferences.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/flux-1-fill-dev-onereward.jpg","ogImage":"https://assets.runware.ai/og/flux-1-fill-dev-onereward.png"},{"model":"bfl-flux-1-expand-pro","air":"bfl:1@3","name":"FLUX.1 Expand [pro]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2024-11-21T00:00:00Z","weight":1,"headline":"High fidelity FLUX outpainting for seamless image expansion","description":"FLUX.1 Expand [pro] is an outpainting model that extends images beyond their original frame while preserving structure, lighting, and style. It supports controlled expansion from real or generated inputs and integrates into image editing or generative workflows that need precise, coherent borders.","capabilities":["io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-expand-pro.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-expand-pro.png","pricingMeasured":[{"configuration":"1K expansion · 50 steps","price":0.05,"latencyMs":22889},{"configuration":"0.5K expansion · 50 steps","price":0.05,"latencyMs":30636},{"configuration":"2K expansion · 50 steps","price":0.05,"latencyMs":23607},{"configuration":"4K expansion · 50 steps","price":0.05,"latencyMs":22028}],"pricingExamples":[{"configuration":"expand","price":"$0.05","amount":0.05,"latencyMs":13684,"exampleId":"monsoon-observatory-campaign-banner"},{"configuration":"expand","price":"$0.05","amount":0.05,"latencyMs":14720,"exampleId":"meteorite-conservation-magazine-opener"},{"configuration":"expand","price":"$0.05","amount":0.05,"latencyMs":14467,"exampleId":"avalanche-patrol-team-portrait"},{"configuration":"expand","price":"$0.05","amount":0.05,"latencyMs":14171,"exampleId":"pipe-organ-restoration-editorial-panorama"}]},{"model":"flux-1-dev-ip-adapter-xlabs-v4","air":"runware:56@4","name":"FLUX.1 [dev] IP-Adapter (XLabs) v4","status":"live","releasedAt":"2024-11-15T00:00:00Z","weight":1,"headline":"Newest exposed XLabs FLUX.1 [dev] IP-Adapter checkpoint in the versioned adapter family","description":"FLUX.1 [dev] IP-Adapter (XLabs) v4 is the latest exposed checkpoint in this versioned XLabs FLUX IP-Adapter family. It represents another selectable iteration of FLUX.1 [dev] image-reference conditioning for teams that want to target a specific checkpoint generation inside the same adapter line.","architecture":"flux-1-dev","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-ip-adapter-xlabs-v4.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-ip-adapter-xlabs-v4.png"},{"model":"bfl-flux-1-1-pro-ultra","air":"bfl:2@2","name":"FLUX.1.1 [pro] Ultra","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2024-11-06T00:00:00Z","weight":1,"headline":"High speed 4MP FLUX image generation for production apps","description":"FLUX.1.1 [pro] Ultra is a high resolution text to image model from Black Forest Labs. It generates images up to 4 megapixels in about 10 seconds. Ultra mode targets sharp outputs. Raw mode targets natural photographic style. Built for API integration in real products.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-1-pro-ultra.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-1-pro-ultra.png","pricingRates":[{"amount":0.06,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.06","amount":0.06,"latencyMs":9951,"exampleId":"bioluminescent-kelp-farm-wall-art"},{"configuration":"text-to-image","price":"$0.06","amount":0.06,"latencyMs":9823,"exampleId":"overnight-transplant-courier-illustration"},{"configuration":"text-to-image","price":"$0.06","amount":0.06,"latencyMs":25833,"exampleId":"converted-bowling-alley-office-interior"},{"configuration":"text-to-image","price":"$0.06","amount":0.06,"latencyMs":11631,"exampleId":"runway-rubber-removal-video-thumbnail"}]},{"model":"bfl-flux-1-1-pro","air":"bfl:2@1","name":"FLUX.1.1 [pro]","creator":"black-forest-labs","zdr":true,"status":"live","releasedAt":"2024-10-02T00:00:00Z","weight":1,"headline":"FLUX.1.1 Pro high fidelity text to image generation","description":"FLUX.1.1 Pro is a flagship text to image model from Black Forest Labs. It improves on FLUX.1 with sharper detail, stronger prompt adherence, and faster sampling. Ideal for production image pipelines, product visuals, and creative tools that require consistent high quality output.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-1-pro.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-1-pro.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":6013,"exampleId":"subsea-cable-engineer-editorial-portrait"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":6997,"exampleId":"oyster-hatchery-risograph-editorial"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":5158,"exampleId":"dung-beetle-navigation-illustration"},{"configuration":"text-to-image","price":"$0.04","amount":0.04,"latencyMs":6504,"exampleId":"anechoic-lab-executive-portrait"}]},{"model":"klingai-lip-sync","air":"klingai:7@1","name":"KlingAI Lip-Sync","creator":"klingai","status":"live","releasedAt":"2024-10-01T00:00:00Z","weight":1,"headline":"Accurate AI lip sync for character driven video content","description":"KlingAI Lip-Sync aligns mouth motion and facial expression with new dialogue or music in existing video. Upload Kling generated clips or compatible footage, attach an audio track, then get back natural synced performance that fits multi character scenes and production workflows.","capabilities":["io:video-to-video","io:audio-to-video","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/klingai-lip-sync.jpg","ogImage":"https://assets.runware.ai/og/klingai-lip-sync.png","pricingOverview":"$0.0462 for up to 5s, $0.0924 for up to 10s, then $0.0092 per additional second.","pricingRates":[{"amount":0.0462,"unit":"output","label":"up to 5s of audio"},{"amount":0.0924,"unit":"output","label":"up to 10s of audio"},{"amount":0.0092,"unit":"durationSecond","after":10,"label":"each second beyond ten"}],"pricingExamples":[{"configuration":"audio-driven","price":"$0.32","amount":0.3202584,"latencyMs":153439,"exampleId":"forensic-botany-series-teaser"},{"configuration":"audio-driven","price":"$0.32","amount":0.3202584,"latencyMs":154584,"exampleId":"aquaculture-oxygen-meter-explainer-clip"},{"configuration":"audio-driven","price":"$0.32","amount":0.3202584,"latencyMs":132837,"exampleId":"planetarium-midnight-broadcast-ident"},{"configuration":"audio-driven","price":"$0.32","amount":0.3202584,"latencyMs":159535,"exampleId":"seaweed-rain-jacket-showcase-reel"}]},{"model":"flux-1-dev-controlnet-union-pro-canny","air":"runware:25@1","name":"FLUX.1 [dev] ControlNet Union Pro – Canny","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":1,"headline":"FLUX.1 ControlNet mode for locking composition and contours from Canny edge maps","description":"FLUX.1 ControlNet Union Pro – Canny is the Canny-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is built for workflows where edge maps should guide composition, silhouette, and scene structure while the prompt controls final style, texture, and subject detail.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-canny.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-canny.png"},{"model":"flux-1-dev-controlnet-union-pro-depth","air":"runware:27@1","name":"FLUX.1 [dev] ControlNet Union Pro – Depth","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":1,"headline":"FLUX.1 ControlNet mode for preserving scene geometry from depth maps","description":"FLUX.1 ControlNet Union Pro – Depth is the depth-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is built for workflows where camera perspective, relative object depth, and large-scale spatial structure need to stay aligned with a source depth map while the prompt reshapes the final image.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-depth.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-depth.png"},{"model":"flux-1-dev-controlnet-union-pro-tile","air":"runware:26@1","name":"FLUX.1 [dev] ControlNet Union Pro – Tile","status":"live","releasedAt":"2024-08-26T00:00:00Z","weight":1,"headline":"FLUX.1 ControlNet mode for tile-guided detail enhancement and structured upscaling","description":"FLUX.1 ControlNet Union Pro – Tile is the tile-conditioned mode of the unified FLUX.1 ControlNet Union Pro model. It is suited to detail enhancement, super-resolution-style refinement, and controlled image regeneration where local texture and structure from the source image should stay anchored.","architecture":"flux-1-dev","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/flux-1-dev-controlnet-union-pro-tile.jpg","ogImage":"https://assets.runware.ai/og/flux-1-dev-controlnet-union-pro-tile.png"},{"model":"ideogram-2-0","air":"ideogram:3@1","name":"Ideogram 2.0","creator":"ideogram","zdr":true,"status":"live","releasedAt":"2024-08-21T00:00:00Z","weight":1,"headline":"Ideogram 2.0 text to image model for sharp design work","description":"Ideogram 2.0 is a frontier text to image model with strong typography control, improved rendering quality, and better layout consistency. It suits branding workflows, posters, and production design where legible stylized text and precise graphic composition matter to developers.","capabilities":["io:text-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/ideogram-2-0.jpg","ogImage":"https://assets.runware.ai/og/ideogram-2-0.png","pricingOverview":"Each generation costs $0.08.","pricingRates":[{"amount":0.05,"unit":"output","label":"Turbo"},{"amount":0.08,"unit":"output","label":"Default"}],"pricingExamples":[{"configuration":"text-to-image","price":"$0.08","amount":0.08,"latencyMs":23191,"exampleId":"geothermal-educator-profile-avatar"},{"configuration":"text-to-image","price":"$0.08","amount":0.08,"latencyMs":22254,"exampleId":"museum-cafe-dessert-menu-cover"}]},{"model":"bfl-flux-1-dev","air":"runware:101@1","name":"FLUX.1 [dev]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-08-01T00:00:00Z","weight":1,"headline":"Open-weight 12B text to image model for rich visuals","description":"FLUX.1 [dev] is a 12-billion-parameter text-to-image model from Black Forest Labs. Its guidance-distilled architecture combines strong prompt adherence with detailed, high-quality visual generation, making it a versatile foundation for creative image workflows and customized model development.","architecture":"flux-1-dev","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-dev.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-dev.png","pricingOverview":"Each image generation costs $0.0038 at 1024x1024.","pricingMeasured":[{"configuration":"1024x1024 · 28 steps","price":0.0038},{"configuration":"512x512 · 28 steps","price":0.0019},{"configuration":"1024x1024 · 28 steps +1 LoRA","price":0.0051},{"configuration":"1024x1024 · 28 steps +2 LoRAs","price":0.058}],"pricingExamples":[{"configuration":"text-to-image · 1024×1024 · steps 28","price":"$0.0013","amount":0.0013,"latencyMs":2942,"exampleId":"coastal-radar-station-rental-interior"},{"configuration":"expand · 1536×1024","price":"$0.0032","amount":0.0032,"latencyMs":7612,"exampleId":"bespoke-sign-painter-portrait"},{"configuration":"image-to-image · 720p","price":"$0.0038","amount":0.0038,"latencyMs":4263,"exampleId":"wastewater-science-picture-book-spread"},{"configuration":"inpaint · 1024×1024","price":"$0.0038","amount":0.0038,"latencyMs":8294,"exampleId":"desert-survival-save-station-concept"}]},{"model":"bfl-flux-1-schnell","air":"runware:100@1","name":"FLUX.1 [schnell]","creator":"black-forest-labs","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-08-01T00:00:00Z","weight":1,"headline":"Ultra fast FLUX.1 text to image model for local use","description":"FLUX.1 [schnell] is an open source text to image model from Black Forest Labs. It uses 4 step distillation for very fast generation with strong visual quality. Ideal for local deployment, rapid prototyping, batch image production, and integration into custom creative pipelines.","architecture":"flux-1-schnell","capabilities":["io:text-to-image","io:image-to-image","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/bfl-flux-1-schnell.jpg","ogImage":"https://assets.runware.ai/og/bfl-flux-1-schnell.png","pricingOverview":"~$0.0018 per megapixel","pricingMeasured":[{"configuration":"512x512 · 4 steps","price":0.0006},{"configuration":"1024x1024 · 4 steps","price":0.0013},{"configuration":"512x512 · 4 steps +1 LoRA","price":0.0006},{"configuration":"1024x1024 · 4 steps +2 LoRAs","price":0.0013}],"pricingExamples":[{"configuration":"image-to-image · 1024×768","price":"$0.0006","amount":0.0006,"latencyMs":3803,"exampleId":"modernist-lido-fundraiser-invitation"},{"configuration":"inpaint · 1024×1024","price":"$0.0006","amount":0.0006,"latencyMs":4161,"exampleId":"forensic-botany-book-cover"},{"configuration":"text-to-image · 720p","price":"$0.0006","amount":0.0006,"latencyMs":1233,"exampleId":"rooftop-beekeeper-editorial-headshot"},{"configuration":"expand · 1536×1024","price":"$0.0013","amount":0.0013,"latencyMs":15455,"exampleId":"danube-delta-tourism-campaign-panorama"}]},{"model":"birefnet-portrait","air":"runware:112@10","name":"BiRefNet Portrait","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-02-12T00:00:00Z","weight":1,"headline":"Human portrait segmentation with refined boundary accuracy","description":"BiRefNet Portrait is tuned for human subject segmentation, focusing on accurate extraction of faces and bodies with clean contours for portrait-centric workflows.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-portrait.jpg","ogImage":"https://assets.runware.ai/og/birefnet-portrait.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5015}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5138,"exampleId":"planetarium-educator-background-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5500,"exampleId":"aluminized-foundry-suit-catalog-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":7021,"exampleId":"wheelchair-racer-campaign-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":9786,"exampleId":"horologist-conference-portrait-cutout"}]},{"model":"birefnet-dis","air":"runware:112@3","name":"BiRefNet Dis","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"Dichotomous image segmentation using bilateral reference modeling","description":"BiRefNet Dis focuses on binary object segmentation tasks where clear separation between subject and background is required. The model emphasizes structural consistency and minimizes foreground leakage through dual reference guidance.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-dis.jpg","ogImage":"https://assets.runware.ai/og/birefnet-dis.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5187}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6278,"exampleId":"twilight-balloon-relighting-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6122,"exampleId":"chainmail-coif-virtual-try-on-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5810,"exampleId":"resort-parasol-colorway-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5345,"exampleId":"panoramic-ballet-campaign-expansion"}]},{"model":"birefnet-general-resolution-512x512-fp16","air":"runware:112@6","name":"BiRefNet General 512×512 FP16","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"Optimized BiRefNet general segmentation at fixed 512×512 resolution","description":"This variant is optimized for efficient inference at a fixed 512×512 resolution using FP16 precision. It balances segmentation quality with lower memory usage and faster runtime for production environments.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-general-resolution-512x512-fp16.jpg","ogImage":"https://assets.runware.ai/og/birefnet-general-resolution-512x512-fp16.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":6200}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":4923,"exampleId":"winter-agritourism-relighting-subject-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5560,"exampleId":"plankton-sampling-net-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5073,"exampleId":"underwater-inspection-drone-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5085,"exampleId":"geothermal-engineer-speaker-headshot"}]},{"model":"birefnet-hrsod-dhu","air":"runware:112@7","name":"BiRefNet HRSOD DHU","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"High-resolution salient object detection with BiRefNet","description":"BiRefNet HRSOD DHU is designed for salient object detection in high-resolution imagery. It preserves fine details and sharp boundaries, making it suitable for complex scenes with small or visually dominant subjects.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-hrsod-dhu.jpg","ogImage":"https://assets.runware.ai/og/birefnet-hrsod-dhu.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5011}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5667,"exampleId":"twilight-hotel-chandelier-relight-asset"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6166,"exampleId":"wind-technician-campaign-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5847,"exampleId":"panoramic-fossil-exhibition-banner"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6761,"exampleId":"moon-jellyfish-exhibit-composite-cutout"}]},{"model":"birefnet-massive-tr-dis5k-tr-tes","air":"runware:112@8","name":"BiRefNet Massive TR DIS5K TR TES","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"Large-scale BiRefNet model trained on diverse segmentation datasets","description":"This model is trained on multiple large-scale datasets, including DIS5K, to improve robustness across object scales, backgrounds, and visual distributions. It prioritizes generalization and consistency in real-world imagery.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-massive-tr-dis5k-tr-tes.jpg","ogImage":"https://assets.runware.ai/og/birefnet-massive-tr-dis5k-tr-tes.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5174}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5836,"exampleId":"couture-fascinator-virtual-try-on-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6791,"exampleId":"night-observatory-relighting-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":4263,"exampleId":"wardian-terrarium-interior-placement-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5990,"exampleId":"clear-electric-violin-product-cutout"}]},{"model":"birefnet-v1-base","air":"runware:112@1","name":"BiRefNet v1 Base","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"Foundation BiRefNet model for high-accuracy binary image segmentation","description":"BiRefNet v1 Base is the core model of the BiRefNet family, designed for dichotomous image segmentation. It produces clean foreground–background masks with strong boundary precision by leveraging bilateral reference cues that preserve spatial structure and object integrity across diverse scenes.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-v1-base.jpg","ogImage":"https://assets.runware.ai/og/birefnet-v1-base.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5045}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5868,"exampleId":"glass-condenser-catalog-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5532,"exampleId":"winter-lights-topiary-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6846,"exampleId":"lighthouse-keeper-portrait-repair"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":5547,"exampleId":"winter-carnival-horse-relight-cutout"}]},{"model":"birefnet-v1-base-cod","air":"runware:112@2","name":"BiRefNet v1 Base – COD","hosted":true,"zdr":true,"status":"live","releasedAt":"2024-01-07T00:00:00Z","weight":1,"headline":"Camouflaged object detection with bilateral reference segmentation","description":"This BiRefNet variant is specialized for camouflaged object detection, targeting objects that closely blend with their surroundings. It enhances subtle boundary contrast and improves detection reliability in low-visibility and texture-confusing scenarios.","capabilities":["io:image-to-image","op:remove-background","form:checkpoint","feat:transparency"],"coverImage":"https://assets.runware.ai/covers/birefnet-v1-base-cod.jpg","ogImage":"https://assets.runware.ai/og/birefnet-v1-base-cod.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":5676}],"pricingExamples":[{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6501,"exampleId":"alpine-snowcat-twilight-relight-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":6148,"exampleId":"barrel-sauna-seasonal-relight-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":7391,"exampleId":"wetland-auditor-profile-portrait-cutout"},{"configuration":"remove-background","price":"$0.0006","amount":0.0006,"latencyMs":8292,"exampleId":"field-recordist-editorial-cutout"}]},{"model":"sd-1-5-ip-adapter-full-face","air":"runware:55@9","name":"IP-Adapter SD 1.5 Full-Face","status":"live","releasedAt":"2023-11-10T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 IP-Adapter for fuller face-preserving portrait and identity workflows","description":"IP-Adapter SD 1.5 Full-Face is an updated face-oriented variant in the SD 1.5 IP-Adapter family. It is aimed at portrait workflows where fuller facial context, identity retention, and more stable face-driven conditioning matter more than general image prompting alone.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-full-face.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-full-face.png"},{"model":"sdxl-ip-adapter-plus","air":"runware:55@2","name":"IP-Adapter SDXL Plus","status":"live","releasedAt":"2023-09-08T00:00:00Z","weight":1,"headline":"Stable Diffusion XL IP-Adapter with patch-based conditioning for stronger reference adherence","description":"IP-Adapter SDXL Plus is the fine-grained SDXL image-prompt adapter. It uses patch image embeddings to stay closer to the source reference than the base SDXL adapter, making it better for workflows where style, layout cues, or subject appearance need tighter alignment.","architecture":"sdxl","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sdxl-ip-adapter-plus.jpg","ogImage":"https://assets.runware.ai/og/sdxl-ip-adapter-plus.png"},{"model":"sdxl-ip-adapter-vit-h","air":"runware:55@4","name":"IP-Adapter SDXL ViT-H","status":"live","releasedAt":"2023-09-08T00:00:00Z","weight":1,"headline":"Stable Diffusion XL IP-Adapter variant using the lighter OpenCLIP ViT-H encoder","description":"IP-Adapter SDXL ViT-H is the updated SDXL adapter variant trained with the OpenCLIP ViT-H image encoder instead of the larger ViT-bigG backbone. It is intended for SDXL image-prompt workflows that want similar conditioning behavior with lower memory demands at inference time.","architecture":"sdxl","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sdxl-ip-adapter-vit-h.jpg","ogImage":"https://assets.runware.ai/og/sdxl-ip-adapter-vit-h.png"},{"model":"sdxl-controlnet-canny","air":"runware:20@1","name":"SDXL ControlNet Canny","status":"live","releasedAt":"2023-09-07T00:00:00Z","weight":1,"headline":"SDXL ControlNet for edge-guided composition and contour-preserving generation","description":"SDXL ControlNet Canny is a Canny-conditioned ControlNet trained on Stable Diffusion XL Base 1.0. It is used when edge maps should tightly control composition and silhouette while SDXL fills in detail, lighting, and final visual style.","architecture":"sdxl","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sdxl-controlnet-canny.jpg","ogImage":"https://assets.runware.ai/og/sdxl-controlnet-canny.png"},{"model":"sd-1-5-ip-adapter-plus-face","air":"runware:55@8","name":"IP-Adapter SD 1.5 Plus-Face","status":"live","releasedAt":"2023-08-30T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 IP-Adapter for face-guided generation with stronger facial reference fidelity","description":"IP-Adapter SD 1.5 Plus-Face is the face-focused variant of the SD 1.5 Plus adapter. It uses cropped face conditioning so portraits, avatars, and character generations can stay closer to facial features and identity cues from the reference image while still allowing text-driven styling.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-plus-face.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-plus-face.png"},{"model":"alibaba-qwen2-5-vl-7b-instruct","air":"runware:152@2","name":"Qwen2.5-VL-7B-Instruct","creator":"alibaba","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-08-24T00:00:00Z","weight":1,"headline":"Instruction-tuned multimodal vision-language model","description":"Qwen2.5-VL-7B-Instruct is a multimodal model that processes images and text together to perform visual reasoning, captioning, question answering, and structured output generation. It integrates a vision encoder with a 7B instruction-tuned language backbone to support rich interactive multimodal understanding.","capabilities":["io:image-to-text","op:caption","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/alibaba-qwen2-5-vl-7b-instruct.jpg","ogImage":"https://assets.runware.ai/og/alibaba-qwen2-5-vl-7b-instruct.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":5315}],"pricingExamples":[{"configuration":"caption","price":"$0.0013","amount":0.0013,"latencyMs":5569,"exampleId":"culinary-magazine-photo-caption"},{"configuration":"caption","price":"$0.0013","amount":0.0013,"latencyMs":5614,"exampleId":"auction-furniture-lot-caption"},{"configuration":"caption","price":"$0.0019","amount":0.0019,"latencyMs":6724,"exampleId":"vehicle-damage-intake-caption"},{"configuration":"caption","price":"$0.0032","amount":0.0032,"latencyMs":8648,"exampleId":"wildfire-response-editorial-photo-caption"}]},{"model":"sd-1-5-ip-adapter-plus","air":"runware:55@6","name":"IP-Adapter SD 1.5 Plus","status":"live","releasedAt":"2023-08-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 IP-Adapter with patch-level image features for stronger reference fidelity","description":"IP-Adapter SD 1.5 Plus is the fine-grained variant of the Stable Diffusion 1.5 IP-Adapter family. It uses patch-based image features instead of only global embeddings, making it better suited when the output should stay closer to the structure, appearance, or styling cues of a reference image.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-plus.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-plus.png"},{"model":"sd-1-5-ip-adapter-base","air":"runware:55@5","name":"IP-Adapter SD 1.5 Base","status":"live","releasedAt":"2023-08-16T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 IP-Adapter for image-prompt conditioning with lightweight reference guidance","description":"IP-Adapter SD 1.5 Base is the original image-prompt adapter for the Stable Diffusion 1.5 family. It uses a lightweight decoupled image-conditioning pathway so reference images can guide style, subject identity, and overall visual intent without replacing text prompting.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-base.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-base.png"},{"model":"sd-1-5-ip-adapter-light","air":"runware:55@7","name":"IP-Adapter SD 1.5 Light","status":"live","releasedAt":"2023-08-16T00:00:00Z","weight":1,"headline":"Lighter Stable Diffusion 1.5 IP-Adapter tuned for softer image conditioning and prompt compatibility","description":"IP-Adapter SD 1.5 Light is the lighter-weight image-prompt adapter in the Stable Diffusion 1.5 family. It is designed to preserve better compatibility with text prompts, making it useful when a reference image should influence the output without dominating the generation as strongly as higher-fidelity variants.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-light.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-light.png"},{"model":"sd-1-5-ip-adapter-vit-h","air":"runware:55@10","name":"IP-Adapter SD 1.5 ViT-H","status":"live","releasedAt":"2023-08-16T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 IP-Adapter variant built around OpenCLIP ViT-H image conditioning","description":"IP-Adapter SD 1.5 ViT-H is a Stable Diffusion 1.5 image-prompt adapter variant centered on the OpenCLIP ViT-H image encoder. It is useful when the deployment exposes the encoder-backed adapter as a separate endpoint and image references should guide the output through the standard IP-Adapter conditioning path.","architecture":"sd-1-5","capabilities":["form:ip-adapter"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-ip-adapter-vit-h.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-ip-adapter-vit-h.png"},{"model":"sd-1-5-controlnet-br-v1","air":"runware:6@2","name":"SD 1.5 ControlNet br v1","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet variant for brightness-guided image control","description":"SD 1.5 ControlNet br v1 is a brightness-oriented ControlNet variant in the Stable Diffusion 1.5 family. It is intended for workflows where coarse luminance structure or brightness distribution should influence how the final image is reconstructed and stylized.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-br-v1.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-br-v1.png"},{"model":"sd-1-5-controlnet-canny-ip2p","air":"runware:6@1","name":"SD 1.5 ControlNet Canny IP2P","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for instruction-based image editing with stronger structural retention","description":"SD 1.5 ControlNet Canny IP2P combines instruction-following image editing with stronger structural control in the Stable Diffusion 1.5 family. It is suited to workflows where edits should follow text instructions while staying closer to the source image's composition and edges.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-canny-ip2p.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-canny-ip2p.png"},{"model":"sd-1-5-controlnet-inpaint","air":"runware:11@1","name":"SD 1.5 ControlNet Inpaint","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for masked inpainting with structural guidance","description":"SD 1.5 ControlNet Inpaint is the inpainting-conditioned ControlNet in the Stable Diffusion 1.5 family. It is designed for masked editing workflows where the unmasked context should stay stable while the controlled region is regenerated with prompt guidance.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-inpaint.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-inpaint.png"},{"model":"sd-1-5-controlnet-lineart","air":"runware:12@1","name":"SD 1.5 ControlNet Lineart","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for line-art-driven composition and stylized generation","description":"SD 1.5 ControlNet Lineart is the lineart-conditioned ControlNet in the Stable Diffusion 1.5 family. It is intended for workflows where clean line drawings or extracted outlines should guide composition while the model fills in color, texture, and style.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-lineart.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-lineart.png"},{"model":"sd-1-5-controlnet-mlsd","air":"runware:13@1","name":"SD 1.5 ControlNet MLSD","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for straight-line and architectural structure guidance","description":"SD 1.5 ControlNet MLSD is the MLSD-conditioned ControlNet in the Stable Diffusion 1.5 family. It is built for layouts dominated by straight edges and line segments, such as architecture, interiors, packaging geometry, diagrams, and other structured compositions.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-mlsd.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-mlsd.png"},{"model":"sd-1-5-controlnet-segmentation","air":"runware:17@1","name":"SD 1.5 ControlNet Segmentation","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for region-aware generation from segmentation maps","description":"SD 1.5 ControlNet Segmentation is the segmentation-conditioned ControlNet in the Stable Diffusion 1.5 family. It is useful when large semantic regions should stay organized according to a segmentation map while the model generates the final scene detail and style.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-segmentation.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-segmentation.png"},{"model":"sd-1-5-controlnet-tile","air":"runware:142@0","name":"SD 1.5 ControlNet Tile","status":"live","releasedAt":"2023-06-23T00:00:00Z","weight":1,"headline":"Stable Diffusion 1.5 ControlNet for tile-guided detail enhancement and super-resolution-style refinement","description":"SD 1.5 ControlNet Tile is the tile-conditioned ControlNet in the Stable Diffusion 1.5 family. It is intended for image enhancement and detail-generation workflows where the source image should guide local structure, often in upscale-like or same-size refinement tasks.","architecture":"sd-1-5","capabilities":["form:controlnet"],"coverImage":"https://assets.runware.ai/covers/sd-1-5-controlnet-tile.jpg","ogImage":"https://assets.runware.ai/og/sd-1-5-controlnet-tile.png"},{"model":"stable-diffusion-latent-upscaler","air":"runware:502@1","name":"Stable Diffusion Latent Upscaler","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-03-23T00:00:00Z","weight":1,"headline":"2× latent-space upscaler for Stable Diffusion outputs","description":"Stable Diffusion Latent Upscaler is a diffusion upscaling model that operates in Stable Diffusion’s latent space to perform fast 2× resolution enhancement before VAE decoding, improving detail without changing global composition.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/stable-diffusion-latent-upscaler.jpg","ogImage":"https://assets.runware.ai/og/stable-diffusion-latent-upscaler.png","pricingMeasured":[{"configuration":"upscaleFactor 2 · settings.steps 20","price":0.0026,"latencyMs":10663},{"configuration":"upscaleFactor 2 · settings.steps 60","price":0.0045,"latencyMs":15268}],"pricingExamples":[{"configuration":"upscale","price":"$0.0032","amount":0.0032,"latencyMs":13670,"exampleId":"railway-safety-feature-illustration"},{"configuration":"upscale","price":"$0.0032","amount":0.0032,"latencyMs":11031,"exampleId":"marionette-workshop-photo-restoration"},{"configuration":"upscale","price":"$0.0045","amount":0.0045,"latencyMs":13295,"exampleId":"antique-diving-helmet-catalog-cutout"},{"configuration":"upscale","price":"$0.0058","amount":0.0058,"latencyMs":14355,"exampleId":"aerospace-recruitment-portrait-composite"}]},{"model":"yolov8n-hand","air":"runware:35@3","name":"YOLOv8n Hand","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-01-10T00:00:00Z","weight":1,"headline":"Lightweight hand mask generation for realistic images","description":"YOLOv8n Hand is a lightweight image masking model for detecting hands in realistic 2D imagery and returning black-and-white masks for each detection. It is useful for automated hand-focused inpainting, hand detail enhancement, and batch masking workflows.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/yolov8n-hand.jpg","ogImage":"https://assets.runware.ai/og/yolov8n-hand.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":4456}],"pricingExamples":[{"configuration":"hand-mask","price":"$0.0006","amount":0.0006,"latencyMs":3768,"exampleId":"cabin-twilight-relight-hand-masks"},{"configuration":"hand-mask","price":"$0.0013","amount":0.0013,"latencyMs":4859,"exampleId":"antarctic-radio-archive-hand-masks"},{"configuration":"hand-mask","price":"$0.0013","amount":0.0013,"latencyMs":5281,"exampleId":"avant-garde-jewelry-hand-cutouts"},{"configuration":"hand-mask","price":"$0.0013","amount":0.0013,"latencyMs":5444,"exampleId":"seaweed-herbarium-hand-cleanup"}]},{"model":"yolov8n-person-seg","air":"runware:35@4","name":"YOLOv8n Person Seg","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-01-10T00:00:00Z","weight":1,"headline":"Lightweight person segmentation mask generation","description":"YOLOv8n Person Seg is a lightweight person segmentation model that generates black-and-white masks for detected people in an image. It is useful for subject-isolated inpainting, person-targeted enhancement, and automated mask generation where speed and low compute cost matter.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/yolov8n-person-seg.jpg","ogImage":"https://assets.runware.ai/og/yolov8n-person-seg.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":4174}],"pricingExamples":[{"configuration":"person-mask","price":"$0.0006","amount":0.0006,"latencyMs":3458,"exampleId":"regency-costume-virtual-try-on-mask"},{"configuration":"person-mask","price":"$0.0006","amount":0.0006,"latencyMs":3979,"exampleId":"theremin-composer-press-headshot"},{"configuration":"person-mask","price":"$0.0013","amount":0.0013,"latencyMs":5251,"exampleId":"desert-dance-festival-composite"},{"configuration":"person-mask","price":"$0.0013","amount":0.0013,"latencyMs":5079,"exampleId":"planetarium-presenter-background-cutout"}]},{"model":"yolov8s-face","air":"runware:35@2","name":"YOLOv8s Face","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-01-10T00:00:00Z","weight":1,"headline":"Higher-accuracy face mask generation for realistic images","description":"YOLOv8s Face is a face-focused image masking model that detects realistic 2D faces and produces black-and-white masks for each matched region. Compared with smaller variants, it is better suited to workflows that need stronger detection accuracy and cleaner face coverage.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/yolov8s-face.jpg","ogImage":"https://assets.runware.ai/og/yolov8s-face.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3416}],"pricingExamples":[{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4986,"exampleId":"observatory-blue-hour-relight-face-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4030,"exampleId":"drag-revue-marquee-outpaint-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4289,"exampleId":"apiary-banner-outpaint-protection-mask"},{"configuration":"face-mask","price":"$0.0019","amount":0.0019,"latencyMs":5551,"exampleId":"polar-night-expedition-face-masks"}]},{"model":"yolov8s-person-seg","air":"runware:35@5","name":"YOLOv8s Person Seg","hosted":true,"zdr":true,"status":"live","releasedAt":"2023-01-10T00:00:00Z","weight":1,"headline":"Higher-precision person segmentation mask generation","description":"YOLOv8s Person Seg is a person segmentation model that produces black-and-white masks for detected human subjects with stronger precision than smaller variants. It is better suited to workflows that need cleaner subject contours for inpainting, compositing, or detail enhancement.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/yolov8s-person-seg.jpg","ogImage":"https://assets.runware.ai/og/yolov8s-person-seg.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3658}],"pricingExamples":[{"configuration":"person-mask","price":"$0.0006","amount":0.0006,"latencyMs":4649,"exampleId":"artistic-swimmer-campaign-cutout"},{"configuration":"person-mask","price":"$0.0006","amount":0.0006,"latencyMs":4300,"exampleId":"perfumer-portrait-outpaint-protection-mask"},{"configuration":"person-mask","price":"$0.0013","amount":0.0013,"latencyMs":5939,"exampleId":"wildland-crew-member-addition-mask"},{"configuration":"person-mask","price":"$0.0013","amount":0.0013,"latencyMs":5396,"exampleId":"mariachi-lineup-removal-masks"}]},{"model":"picsart-image-vectorizer","air":"picsart:1@1","name":"Picsart Image Vectorizer","creator":"picsart","status":"live","releasedAt":"2022-02-01T00:00:00Z","weight":1,"headline":"Convert raster graphics into clean scalable SVG vectors","description":"Picsart Image Vectorizer converts raster images into precise SVG vectors. It performs edge aware tracing and separates flat color regions. Ideal for developers who need compact, resolution independent assets for logos, icons, UI elements, and flat illustrations.","capabilities":["io:image-to-image","op:vectorize","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/picsart-image-vectorizer.jpg","ogImage":"https://assets.runware.ai/og/picsart-image-vectorizer.png","pricingRates":[{"amount":0.04,"unit":"output"}],"pricingExamples":[{"configuration":"vectorize","price":"$0.04","amount":0.04,"latencyMs":17421,"exampleId":"polar-rover-exhibition-backdrop"},{"configuration":"vectorize","price":"$0.04","amount":0.04,"latencyMs":5343,"exampleId":"coral-bioacoustics-laboratory-logo"},{"configuration":"vectorize","price":"$0.04","amount":0.04,"latencyMs":14723,"exampleId":"archaeological-vessel-diagram-cleanup"},{"configuration":"vectorize","price":"$0.04","amount":0.04,"latencyMs":15899,"exampleId":"historic-cable-car-exhibit-scene"}]},{"model":"swinir","air":"runware:503@1","name":"SwinIR","hosted":true,"zdr":true,"status":"live","releasedAt":"2021-08-26T00:00:00Z","weight":1,"headline":"Transformer-based image restoration upscaler for clean, sharp results","description":"SwinIR is an image restoration and super-resolution model based on Swin Transformer blocks, producing high-quality upscales with strong texture recovery and low artifact levels across common restoration settings.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/swinir.jpg","ogImage":"https://assets.runware.ai/og/swinir.png","pricingMeasured":[{"configuration":"upscaleFactor 2","price":0.0032,"latencyMs":8358}],"pricingExamples":[{"configuration":"upscale","price":"$0.0019","amount":0.0019,"latencyMs":7197,"exampleId":"vintage-sextant-auction-catalog-master"},{"configuration":"upscale","price":"$0.0032","amount":0.0032,"latencyMs":10621,"exampleId":"arctic-greenhouse-blue-hour-enlargement"},{"configuration":"upscale","price":"$0.0038","amount":0.0038,"latencyMs":10454,"exampleId":"cyanotype-fossil-exhibition-portrait"},{"configuration":"upscale","price":"$0.007","amount":0.007,"latencyMs":16958,"exampleId":"japanese-firecoat-conservation-master"}]},{"model":"real-esrgan","air":"runware:504@1","name":"Real-ESRGAN","hosted":true,"zdr":true,"status":"live","releasedAt":"2021-07-22T00:00:00Z","weight":1,"headline":"Practical GAN-based upscaler for real-world images","description":"Real-ESRGAN is a deep learning-based image super-resolution model that enhances low-resolution images by inferring high-frequency detail and preserving texture quality. It uses generative adversarial techniques to upscale images while reducing noise and maintaining natural appearance.","capabilities":["io:image-to-image","op:upscale","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/real-esrgan.jpg","ogImage":"https://assets.runware.ai/og/real-esrgan.png","pricingMeasured":[{"configuration":"upscaleFactor 2","price":0.0006,"latencyMs":3983},{"configuration":"upscaleFactor 4","price":0.0013,"latencyMs":7347}],"pricingExamples":[{"configuration":"upscale","price":"$0.0013","amount":0.0013,"latencyMs":6636,"exampleId":"prosthetic-running-blade-catalog-photo"},{"configuration":"upscale","price":"$0.0013","amount":0.0013,"latencyMs":9722,"exampleId":"desert-observatory-night-relight"},{"configuration":"upscale","price":"$0.0026","amount":0.0026,"latencyMs":13497,"exampleId":"saffron-apiary-smoker-colorway"},{"configuration":"upscale","price":"$0.0026","amount":0.0026,"latencyMs":15897,"exampleId":"herbarium-curator-catalogue-portrait"}]},{"model":"openai-clip-vit-l-14","air":"runware:151@1","name":"OpenAI CLIP ViT-L/14","creator":"openai","hosted":true,"zdr":true,"status":"live","releasedAt":"2021-02-26T00:00:00Z","weight":1,"headline":"Vision encoder for text-image representation and similarity","description":"OpenAI CLIP ViT-L/14 is a contrastive vision-language model that embeds images and text into a shared representation space. It enables tasks like zero-shot image classification, semantic search, and similarity scoring by computing aligned feature vectors for images and texts.","capabilities":["io:image-to-text","op:caption","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/openai-clip-vit-l-14.jpg","ogImage":"https://assets.runware.ai/og/openai-clip-vit-l-14.png","pricingMeasured":[{"configuration":"Default","price":0.0032,"latencyMs":7852}],"pricingExamples":[{"configuration":"caption","price":"$0.0038","amount":0.0038,"latencyMs":8383,"exampleId":"museum-exhibition-accessibility-caption"},{"configuration":"caption","price":"$0.0038","amount":0.0038,"latencyMs":9202,"exampleId":"residential-storm-damage-claim-caption"},{"configuration":"caption","price":"$0.0083","amount":0.0083,"latencyMs":14378,"exampleId":"warehouse-safety-audit-caption"},{"configuration":"caption","price":"$0.0237","amount":0.0237,"latencyMs":39636,"exampleId":"aquaculture-editorial-photo-caption"}]},{"model":"mediapipe-eyes-lips-mesh","air":"runware:35@10","name":"MediaPipe Eyes + Lips Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Combined eye and lip mask generation from facial landmarks","description":"MediaPipe Eyes + Lips Mesh is a facial-feature masking model that combines eye-region and lip-region masks into a single black-and-white output. It is useful for selective beauty edits, expression-focused inpainting, and localized detail enhancement workflows.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-eyes-lips-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-eyes-lips-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3925}],"pricingExamples":[{"configuration":"eyes-lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":3899,"exampleId":"museum-trustee-composite-feature-mask"},{"configuration":"eyes-lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":4836,"exampleId":"marine-salvage-crew-insert-mask"},{"configuration":"eyes-lips-mask","price":"$0.0013","amount":0.0013,"latencyMs":5947,"exampleId":"nocturnal-fieldwork-portrait-relight-matte"},{"configuration":"eyes-lips-mask","price":"$0.0013","amount":0.0013,"latencyMs":5981,"exampleId":"opening-night-chef-composite-matte"}]},{"model":"mediapipe-eyes-mesh","air":"runware:35@15","name":"MediaPipe Eyes Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Specialized eye mask generation from facial landmarks","description":"MediaPipe Eyes Mesh is an image masking model specialized for generating black-and-white masks over the eye regions using MediaPipe facial landmarks. It is suited to selective edits, makeup-style retouching, and targeted eye-area enhancement workflows.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-eyes-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-eyes-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":4831}],"pricingExamples":[{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":4089,"exampleId":"planetarium-banner-outpaint-eye-matte"},{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":3921,"exampleId":"vfx-eye-marker-cleanup-matte"},{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":4171,"exampleId":"synchronized-swimmer-eye-cleanup-mask"},{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":4318,"exampleId":"literary-linocut-portrait-restyle"}]},{"model":"mediapipe-face-mesh","air":"runware:35@8","name":"MediaPipe Face Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Landmark-aware face mask generation with MediaPipe mesh topology","description":"MediaPipe Face Mesh is a face masking model built on dense facial landmark topology. It produces black-and-white masks with more contour awareness than simple bounding-box detectors, making it useful for precision face inpainting, facial retouching, and detail enhancement pipelines.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-face-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-face-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3090}],"pricingExamples":[{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3686,"exampleId":"planetarium-director-headshot-retouch-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3666,"exampleId":"emergency-interpreter-composite-face-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4068,"exampleId":"roller-derby-archive-face-mask"},{"configuration":"face-mask","price":"$0.0013","amount":0.0013,"latencyMs":5156,"exampleId":"opera-key-art-face-holdout"}]},{"model":"mediapipe-face-mesh-eyes-only","air":"runware:35@9","name":"MediaPipe Face Mesh Eyes Only","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Eye-region mask generation from face mesh landmarks","description":"MediaPipe Face Mesh Eyes Only is a feature-targeted image masking model that isolates the eye regions from MediaPipe face landmarks and returns black-and-white masks. It is useful for selective inpainting, beauty workflows, stylization, and localized eye-area enhancement.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-face-mesh-eyes-only.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-face-mesh-eyes-only.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":4505}],"pricingExamples":[{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":3424,"exampleId":"museum-banner-outpaint-protection-matte"},{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":3799,"exampleId":"cenote-expedition-background-swap-matte"},{"configuration":"eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":3662,"exampleId":"seed-bank-curator-retouch-mask"},{"configuration":"eyes-mask","price":"$0.0013","amount":0.0013,"latencyMs":4855,"exampleId":"creature-contact-lens-cleanup-mask"}]},{"model":"mediapipe-lips-mesh","air":"runware:35@14","name":"MediaPipe Lips Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Specialized lip mask generation from facial landmarks","description":"MediaPipe Lips Mesh is a facial-feature masking model that isolates the lips using MediaPipe landmark topology and returns black-and-white masks. It is useful for selective lip edits, makeup transfers, restoration, and targeted inpainting workflows.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-lips-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-lips-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3979}],"pricingExamples":[{"configuration":"lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":4205,"exampleId":"speech-therapy-lip-cutout-mask"},{"configuration":"lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":4223,"exampleId":"ferry-safety-presenter-lip-holdout"},{"configuration":"lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":4140,"exampleId":"volcanology-documentary-mouth-holdout"},{"configuration":"lips-mask","price":"$0.0013","amount":0.0013,"latencyMs":5676,"exampleId":"seasonal-lipstick-colorway-isolation-mask"}]},{"model":"mediapipe-nose-eyes-mesh","air":"runware:35@11","name":"MediaPipe Nose + Eyes Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Combined nose and eye mask generation from facial landmarks","description":"MediaPipe Nose + Eyes Mesh is a facial-feature masking model that generates a combined black-and-white mask for the nose and eye regions. It is useful for portrait retouching, selective facial enhancement, and inpainting tasks that focus on central facial features.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-nose-eyes-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-nose-eyes-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0013,"latencyMs":5517}],"pricingExamples":[{"configuration":"nose-eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":5126,"exampleId":"volcanologist-facial-cleanup-mask"},{"configuration":"nose-eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":5025,"exampleId":"perfumer-cyanotype-restyle-mask"},{"configuration":"nose-eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":7115,"exampleId":"opera-prosthetic-virtual-fitting-mask"},{"configuration":"nose-eyes-mask","price":"$0.0006","amount":0.0006,"latencyMs":5578,"exampleId":"bat-ecologist-dusk-relight-mask"}]},{"model":"mediapipe-nose-lips-mesh","air":"runware:35@12","name":"MediaPipe Nose + Lips Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Combined nose and lip mask generation from facial landmarks","description":"MediaPipe Nose + Lips Mesh is a facial-feature masking model that returns a combined black-and-white mask for the nose and lips. It is useful for selective facial cleanup, retouching, and detail enhancement where lower-face control matters more than full-face masking.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-nose-lips-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-nose-lips-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3954}],"pricingExamples":[{"configuration":"nose-lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":4242,"exampleId":"mars-crew-backdrop-protection-matte"},{"configuration":"nose-lips-mask","price":"$0.0006","amount":0.0006,"latencyMs":5613,"exampleId":"museum-dispatch-rider-repair-matte"},{"configuration":"nose-lips-mask","price":"$0.0019","amount":0.0019,"latencyMs":6662,"exampleId":"risograph-portrait-facial-feature-matte"},{"configuration":"nose-lips-mask","price":"$0.0019","amount":0.0019,"latencyMs":8133,"exampleId":"coral-expedition-cyanotype-facial-matte"}]},{"model":"mediapipe-nose-mesh","air":"runware:35@13","name":"MediaPipe Nose Mesh","hosted":true,"zdr":true,"status":"live","releasedAt":"2020-09-25T00:00:00Z","weight":1,"headline":"Specialized nose-region mask generation from facial landmarks","description":"MediaPipe Nose Mesh is a feature-targeted image masking model that extracts the nose region from MediaPipe face landmarks and outputs black-and-white masks. It is useful for localized retouching, stylized edits, and selective inpainting where the nose area needs isolated control.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-nose-mesh.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-nose-mesh.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3975}],"pricingExamples":[{"configuration":"nose-mask","price":"$0.0006","amount":0.0006,"latencyMs":4749,"exampleId":"volcanologist-portrait-nose-protection-mask"},{"configuration":"nose-mask","price":"$0.0013","amount":0.0013,"latencyMs":5864,"exampleId":"panoramic-perfumer-portrait-expansion"},{"configuration":"nose-mask","price":"$0.0013","amount":0.0013,"latencyMs":5881,"exampleId":"observatory-recruitment-banner-expansion"},{"configuration":"nose-mask","price":"$0.0013","amount":0.0013,"latencyMs":6839,"exampleId":"midnight-chess-portrait-relight-mask"}]},{"model":"mediapipe-face-full","air":"runware:35@6","name":"MediaPipe Face Full","hosted":true,"zdr":true,"status":"live","releasedAt":"2019-07-11T00:00:00Z","weight":1,"headline":"Full-range face mask generation with MediaPipe detection","description":"MediaPipe Face Full is an image masking model for realistic face detection using the full-range MediaPipe face detector. It generates black-and-white masks around detected faces and is useful when the subject may appear farther from the camera or occupy a smaller portion of the frame.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-face-full.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-face-full.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3410}],"pricingExamples":[{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3689,"exampleId":"volcanologist-editorial-headshot-retouch-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3908,"exampleId":"academic-regalia-colorway-exclusion-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4536,"exampleId":"punch-card-lab-restoration-mask"},{"configuration":"face-mask","price":"$0.0013","amount":0.0013,"latencyMs":4810,"exampleId":"ski-goggle-try-on-face-matte"}]},{"model":"mediapipe-face-short","air":"runware:35@7","name":"MediaPipe Face Short","hosted":true,"zdr":true,"status":"live","releasedAt":"2019-07-11T00:00:00Z","weight":1,"headline":"Short-range face mask generation with MediaPipe detection","description":"MediaPipe Face Short is an image masking model optimized for short-range realistic face detection. It returns black-and-white masks around nearby faces and is a practical fit for portrait images, selfies, and other close-frame facial masking workflows.","capabilities":["io:image-to-image","op:mask","form:checkpoint"],"coverImage":"https://assets.runware.ai/covers/mediapipe-face-short.jpg","ogImage":"https://assets.runware.ai/og/mediapipe-face-short.png","pricingMeasured":[{"configuration":"Default","price":0.0006,"latencyMs":3105}],"pricingExamples":[{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3229,"exampleId":"clinical-training-privacy-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3620,"exampleId":"racing-helmet-virtual-try-on-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":4074,"exampleId":"linocut-dance-portrait-face-mask"},{"configuration":"face-mask","price":"$0.0006","amount":0.0006,"latencyMs":3699,"exampleId":"polar-research-banner-expansion-mask"}]},{"model":"ffmpeg","air":"runware:ffmpeg@0","name":"FFmpeg","status":"live","releasedAt":"2000-12-20T00:00:00Z","weight":1,"headline":"Raw FFmpeg processing for video, audio and image files","description":"FFmpeg is a media processing tool that runs one FFmpeg command over files you pass by URL and returns a single output file. It takes up to 10 video, audio or image inputs and accepts raw input options, output options, filter graphs and stream maps. It fits transcoding, trimming, scaling, overlays, audio extraction and format conversion between generation steps.","capabilities":["op:media-process"],"coverImage":"https://assets.runware.ai/covers/ffmpeg.jpg","ogImage":"https://assets.runware.ai/og/ffmpeg.png","pricingRates":[{"amount":0.02,"unit":"gigabyteSecond"}],"pricingExamples":[{"configuration":"media-process","price":"$0.000063","amount":0.000063,"latencyMs":9920,"exampleId":"documentary-interview-dialogue-stem"},{"configuration":"media-process","price":"$0.000099","amount":0.000099,"latencyMs":12116,"exampleId":"perfume-campaign-retouching-still"},{"configuration":"media-process","price":"$0.00328","amount":0.003282,"latencyMs":23688,"exampleId":"thermal-inspection-picture-in-picture-video"},{"configuration":"media-process","price":"$0.00335","amount":0.00335,"latencyMs":23789,"exampleId":"boutique-hotel-social-reel-assembly"}]}]