kinovi.ai

# Kinovi

> One API for video, image, and audio models.

Kinovi runs published video, image, and audio models through a public REST API. POST /api/v1/jobs/createTask with a public model id, then poll recordInfo for status, output URLs, and creditsUsed. Credits are one-time packages at the rates on each model page. Docs include runnable examples. Studio is optional and uses the same balance.

## What Kinovi Is

- Public REST API for video, image, and audio models
- Pay-as-you-go credits, with model IDs and rates published on each model page
- Docs with createTask and recordInfo examples
- Optional Studio in the browser, on the same credit balance

## For API Buyers

- REST API for running video, image, and audio models.
- Task lifecycle uses createTask for submission and recordInfo for polling. Seedance 2.5 drafts upgrade to 1080p with upgradeDraft.
- Pricing is credit-based and model-specific; public model pages show DB-published USD rates when available and credit units otherwise.
- Built for developers who need clear model IDs, request fields, creditsUsed auditing, and stable task polling.
- Each public model page states whether the model is live or announced before presenting API access, pricing, or Studio.

## Key Pages

- [Home](https://kinovi.ai): Product overview for the Kinovi API: video, image, and audio models, with optional Studio.
- [Models](https://kinovi.ai/models): Directory of public AI video, image, and audio model pages with IDs, availability, and published rates.
- [Pricing](https://kinovi.ai/pricing): Per-model starting prices and credit rates.
- [API Reference](https://kinovi.ai/docs/api): Human-readable reference for createTask and recordInfo: authentication, request and response fields, webhooks, error codes, concurrency.
- [Model API docs](https://kinovi.ai/docs/models): Per-model API pages with inputs, pricing, and runnable curl, Python, and TypeScript examples.
- [OpenAPI](https://kinovi.ai/openapi.json): Machine-readable Kinovi public API contract for createTask and recordInfo.
- [MCP server](https://kinovi.ai/api/mcp): Remote Model Context Protocol endpoint (Streamable HTTP) for model discovery, cost estimates, task submission and polling, account balance, and temporary uploads. Supports browser OAuth and bearer API keys.
- [MCP OAuth metadata](https://kinovi.ai/.well-known/oauth-authorization-server): OAuth 2.1 authorization-server metadata for the MCP endpoint, including PKCE, dynamic client registration, and read/write scopes.
- [MCP protected resource metadata](https://kinovi.ai/.well-known/oauth-protected-resource): RFC 9728 metadata binding the MCP resource to its authorization server and scopes.
- [LLM summary](https://kinovi.ai/llms.txt): Concise product and API summary for answer engines.
- [Full LLM index](https://kinovi.ai/llms-full.txt): Full documentation and published model index for answer engines.
- [API catalog](https://kinovi.ai/.well-known/api-catalog): RFC 9727 linkset pointing to the public API contract and health endpoint.
- [AI summary](https://kinovi.ai/ai/summary.json): Short JSON summary of the Kinovi API, primary URLs, and what the API does.
- [AI FAQ](https://kinovi.ai/ai/faq.json): JSON FAQ from the published docs: authentication, credits, concurrency, and retention.
- [AI service descriptor](https://kinovi.ai/ai/service.json): Machine-readable service descriptor: API base, documentation, and bearer authentication.

## Developer Entry Points

- [OpenAPI](https://kinovi.ai/openapi.json): Machine-readable Kinovi public API contract for createTask and recordInfo.
- [Examples on GitHub](https://github.com/kinovi-ai/kinovi-models): Runnable cURL, Node.js, and Python scripts for every model, the same ones shown on each model's docs page.
- [Full LLM index](https://kinovi.ai/llms-full.txt): Full documentation and published model index for answer engines.
- [Create task](https://kinovi.ai/api/v1/jobs/createTask): POST a generation task.
- [Upgrade draft](https://kinovi.ai/api/v1/jobs/upgradeDraft): POST a Seedance 2.5 draft taskId to create a new 1080p task.
- [Poll task](https://kinovi.ai/api/v1/jobs/recordInfo?taskId=TASK_ID): GET task status and generated outputs.
- Authentication: Authorization: Bearer KINOVI_API_KEY

## Common Request Fields

- model: Public model id, for example seedance-20, seedance2-fast, gpt-image-2, or another currently listed public model.
- inputs.prompt: Prompt or generation instruction when the selected model accepts text.
- inputs.urls: Public reference media URLs. The REST API maps this to uploadedUrls internally.
- inputs.imageUrls: Source image URLs for models with an explicit imageUrls field.
- inputs.duration: Video duration in numeric seconds for video models that expose duration.
- inputs.resolution: Aspect-ratio oriented canvas size such as 1280x720 when supported.
- inputs.outputResolution: Native output tier such as 480p, 720p, or 1080p when supported.
- callBackUrl: Optional webhook URL called when the task finishes.

## Documentation

Each page is plain Markdown at the linked `.md` URL (drop `.md` for the HTML page). Directory: https://kinovi.ai/docs.md. Model pages list the exact `createTask` fields, pricing, and runnable cURL / Node.js / Python examples for that model.

- [Account & Credits](https://kinovi.ai/docs/account.md): Read your credit balance and account profile with a single API key request — check funds before createTask.
- [API Reference](https://kinovi.ai/docs/api.md): The Kinovi REST API — authentication, createTask and recordInfo, webhooks, error codes, and concurrency limits.
- [Credits and Billing](https://kinovi.ai/docs/credits-billing.md): How Kinovi charges and refunds credits, how credit packages work, and how API concurrency tiers are assigned.
- [FAQ](https://kinovi.ai/docs/faq.md): Account-wide answers on refunds, credits, concurrency, and what differs between models.
- [MCP Server](https://kinovi.ai/docs/mcp.md): Connect Claude Code, Claude Desktop, Cursor, or any MCP client to Kinovi — discover models, estimate cost, submit tasks, and poll results.
- [API key and custom clients](https://kinovi.ai/docs/mcp/api-key.md): Connect to Kinovi MCP with an API key when OAuth is not available.
- [Build an integration](https://kinovi.ai/docs/mcp/build-an-integration.md): Call Kinovi MCP from your own app on behalf of your users with OAuth.
- [ChatGPT](https://kinovi.ai/docs/mcp/chatgpt.md): Add Kinovi as a custom MCP server in ChatGPT on the web.
- [Claude Code](https://kinovi.ai/docs/mcp/claude-code.md): Add Kinovi MCP to Claude Code and authenticate with OAuth.
- [Claude Desktop](https://kinovi.ai/docs/mcp/claude-desktop.md): Add Kinovi as a custom remote connector in Claude Desktop.
- [Codex CLI](https://kinovi.ai/docs/mcp/codex-cli.md): Add Kinovi MCP to Codex CLI and authenticate with OAuth.
- [Cursor](https://kinovi.ai/docs/mcp/cursor.md): Add Kinovi MCP to Cursor with a remote HTTP server entry.
- [Revoke access](https://kinovi.ai/docs/mcp/disconnect.md): Disconnect an MCP client from your Kinovi account.
- [Privacy and data](https://kinovi.ai/docs/mcp/privacy-and-data.md): What the Kinovi MCP server can access and how tokens are stored.
- [Models](https://kinovi.ai/docs/models.md): Every model on the Kinovi API — request shape, inputs, pricing, and runnable Python, TypeScript, and curl examples.
- [Quickstart](https://kinovi.ai/docs/quickstart.md): Sign in, pick a model, and create a task. Studio and the REST API share one account and credit balance.
- [Uploading Assets](https://kinovi.ai/docs/uploads.md): Two ways to give a model reference images, video, or audio — pass your own public URLs, or upload files to Kinovi storage and use the returned URL.
- [GPT Image 2](https://kinovi.ai/docs/models/gpt-image-2.md): OpenAI's image model on the Kinovi API. Text to image, or edit and restyle existing images with up to 10 references. Strong at photorealism, accurate in-image text, and multi-element compositions.
- [GPT Image 2.5 Flare](https://kinovi.ai/docs/models/gpt-image-2.5-flare.md): OpenAI GPT Image 2.5 Flare on the Kinovi API — the speed-oriented face of GPT Image 2.5. Text to image, or edit with up to 16 reference images. Five quality tiers, 1K–4K output, PNG or JPEG, and usage-based token billing.
- [GPT Image 2.5 Sunburst](https://kinovi.ai/docs/models/gpt-image-2.5-sunburst.md): OpenAI GPT Image 2.5 Sunburst on the Kinovi API. Text to image, or edit with up to 16 reference images. Five quality tiers up to Max, 1K–4K output, PNG or JPEG, and usage-based token billing.
- [Midjourney Niji V7](https://kinovi.ai/docs/models/midjourney-v7-niji.md): Midjourney Niji on the Kinovi API. Anime, manga, and illustration from text, or restyle one reference image. Each task returns four images.
- [Midjourney V8](https://kinovi.ai/docs/models/midjourney-v8.md): Midjourney's image model on the Kinovi API. Text to image, or restyle one reference image. Each task returns four images.
- [Nano Banana Pro](https://kinovi.ai/docs/models/nanobanana-pro.md): Google Gemini Pro image model on the Kinovi API. Text to image, or edit with reference images. Strong at detailed briefs, in-image text, and 1K / 2K / 4K output.
- [NanoBanana 2](https://kinovi.ai/docs/models/nanobanana2.md): Google Gemini image model on the Kinovi API. Text to image, or edit and composite with reference images. Fast generation at 2K or 4K.
- [Seedance 2.0](https://kinovi.ai/docs/models/seedance-20.md): ByteDance's video model on the Kinovi API. Text to video, image to video (single frame or first/last keyframes), or reference-driven generation from up to 9 images, 3 videos and 3 audio clips. 4–15 second clips with a generated audio track, 480p to native 4K.
- [Seedance 2.5](https://kinovi.ai/docs/models/seedance2-5.md): ByteDance's Seedance 2.5 on the Kinovi API. Text to video, image to video (single frame or first/last keyframes), or reference-driven generation from up to 30 images, 10 videos and 10 audio clips. 4–30 second clips with a generated audio track, 480p to 1080p.
- [Seedance 2.5 Video Edit](https://kinovi.ai/docs/models/seedance2-5-edit.md): Edit an existing clip with Seedance 2.5 on the Kinovi API. Pass the video and describe what to change — subject, colours, clothing, background, lighting or style — and the output follows the source's motion, at the source's length and aspect ratio. Source clips of 4–30 seconds, 480p to 1080p.
- [Seedance 2.5 Video Extend](https://kinovi.ai/docs/models/seedance2-5-extend.md): Continue an existing clip with Seedance 2.5 on the Kinovi API. Pass the video, describe what happens next and choose how many seconds to generate — the output is the new footage, starting on the source's last frame, in the source's aspect ratio. 4–30 seconds of new footage, 480p to 1080p.
- [Seedance 2.0 Fast](https://kinovi.ai/docs/models/seedance2-fast.md): ByteDance's Seedance 2.0 tuned for turnaround, on the Kinovi API. Same inputs as Seedance 2.0 — text, images (single frame or first/last keyframes), up to 9 reference images, 3 videos and 3 audio clips — at 480p or 720p and a lower per-second price. 4–15 second clips with a generated audio track.
- [Seedance 2.0 Mini](https://kinovi.ai/docs/models/seedance2.0-mini.md): The smallest, cheapest Seedance 2.0 variant on the Kinovi API — for drafts, batch tests and prompt iteration. Same inputs as Seedance 2.0 (text, keyframe images, up to 9 reference images, 3 videos and 3 audio clips) at 480p or 720p. 4–15 second clips with a generated audio track.
- [Seedream 5.0 Pro](https://kinovi.ai/docs/models/seedream-5.0-pro.md): ByteDance's image model on the Kinovi API. Text to image, or edit existing images with up to 10 references. Strong at controlled edits, layouts, and multilingual in-image text. Output tops out at 2K.
- [Wan 3.0 Image-to-Video](https://kinovi.ai/docs/models/wan3.0-image-to-video.md): Alibaba's Wan 3.0 image-to-video on the Kinovi API. Animate one image as the opening frame, or give a first and a last frame and let the model fill in the motion. 2–30 second clips with a generated audio track, 480p to 1080p.
- [Wan 3.0 Prime Image-to-Video](https://kinovi.ai/docs/models/wan3.0-prime-image-to-video.md): Wan 3.0 Prime image-to-video on the Kinovi API — the faster Wan 3.0 tier. Animate one image as the opening frame, or give a first and a last frame and let the model fill in the motion. 2–30 second clips with a generated audio track, 480p to 1080p.
- [Wan 3.0 Prime Reference-to-Video](https://kinovi.ai/docs/models/wan3.0-prime-ref-to-video.md): Wan 3.0 Prime reference-to-video on the Kinovi API — the faster Wan 3.0 tier. Build a shot from up to 10 reference images, 5 reference videos and 5 audio clips, cited in the prompt as @Image1, @Video1, @Audio1. 2–30 second clips with a generated audio track, 480p to 1080p.
- [Wan 3.0 Prime Text-to-Video](https://kinovi.ai/docs/models/wan3.0-prime-text-to-video.md): Wan 3.0 Prime text-to-video on the Kinovi API — the faster Wan 3.0 tier. Describe a shot and get a clip with a generated audio track. 2–30 seconds, 480p to 1080p, five aspect ratios.
- [Wan 3.0 Reference-to-Video](https://kinovi.ai/docs/models/wan3.0-ref-to-video.md): Alibaba's Wan 3.0 reference-to-video on the Kinovi API. Build a shot from up to 10 reference images, 5 reference videos and 5 audio clips, cited in the prompt as @Image1, @Video1, @Audio1. 2–30 second clips with a generated audio track, 480p to 1080p.
- [Wan 3.0 Text-to-Video](https://kinovi.ai/docs/models/wan3.0-text-to-video.md): Alibaba's Wan 3.0 text-to-video on the Kinovi API. Describe a shot and get a clip with a generated audio track. 2–30 seconds, 480p to 1080p, five aspect ratios.

## Published Model Pages

- A published model page does not by itself claim that Studio, API access, or pricing is live. Check the linked page for current availability.
- [Seedance 2.5](https://kinovi.ai/models/seedance2-5): model ID `seedance2-5`; video; provider ByteDance; status live; rates 480p: $0.1197/s, 480p + reference video: $0.0737/s, 720p: $0.2717/s, 720p + reference video: $0.1612/s, 1080p: $0.6604/s, 1080p + reference video: $0.3947/s, 4K: $1.0964/s, 4K + reference video: $0.6720/s
- [Seedance 2.0](https://kinovi.ai/models/seedance-20): model ID `seedance-20`; video; provider ByteDance; status live; rates 480p: $0.0707/s, 480p + reference video: $0.0895/s, 720p: $0.1884/s, 720p + reference video: $0.2261/s, 1080p: $0.4239/s, 1080p + reference video: $0.5087/s, 4k: $0.9420/s, 4k-reference: $1.1305/s
- [Wan 3.0 Prime Reference-to-Video](https://kinovi.ai/models/wan3-prime-ref-to-video): model ID `wan3.0-prime-ref-to-video`; video; provider Alibaba; status live; rates 480P: $0.0877/s, 720P: $0.1806/s, 1080P: $0.3611/s
- [Wan 3.0 Reference-to-Video](https://kinovi.ai/models/wan3-ref-to-video): model ID `wan3.0-ref-to-video`; video; provider Alibaba; status live; rates 480P: $0.0663/s, 720P: $0.1326/s, 1080P: $0.2651/s
- [Seedream 5.0 Pro](https://kinovi.ai/models/seedream-5-pro): model ID `seedream-5.0-pro`; image; provider ByteDance; status live; rates 1K: $0.0527/image, 2K: $0.0904/image
- [Seedream 5.5 Text to Image](https://kinovi.ai/models/seedream-5-5): model ID `seedream-5.5`; image; provider ByteDance; status live; rates 1k: $0.0180/image, 1.5k: $0.0180/image, 2k: $0.0315/image
- [GPT Image 2](https://kinovi.ai/models/gpt-image-2): model ID `gpt-image-2`; image; provider OpenAI; status live; rates Low · 1K: $0.0100/image, Low · 2K: $0.0200/image, Low · 4K: $0.0300/image, Medium · 1K: $0.0600/image, Medium · 2K: $0.1000/image, Medium · 4K: $0.1800/image, High · 1K: $0.2200/image, High · 2K: $0.4000/image, High · 4K: $0.7200/image
- [Seedance 2.0 Fast](https://kinovi.ai/models/seedance2-fast): model ID `seedance2-fast`; video; provider ByteDance; status live; rates 480p: $0.0495/s, 480p + reference video: $0.0636/s, 720p: $0.0989/s, 720p + reference video: $0.1201/s
- [Seedream 5.5 Image Edit](https://kinovi.ai/models/seedream-5-5-edit): model ID `seedream-5.5-edit`; image; provider ByteDance; status live; rates 1k: $0.0180/image, 1.5k: $0.0180/image, 2k: $0.0315/image, Input image: $0.0072/input image
- [Seedance 2.0 Mini](https://kinovi.ai/models/seedance2-mini): model ID `seedance2.0-mini`; video; provider ByteDance; status live; rates 480p: $0.0212/s, 480p + reference video: $0.0268/s, 720p: $0.0565/s, 720p + reference video: $0.0678/s
- [Seedance 2.5 Video Edit](https://kinovi.ai/models/seedance2-5-edit): model ID `seedance2-5-edit`; video; provider ByteDance; status live; rates 480p-reference: $0.0737/s, 720p-reference: $0.1612/s, 1080p + reference video: $0.3947/s, 4K + reference video: $0.6720/s
- [Seedance 2.5 Video Extend](https://kinovi.ai/models/seedance2-5-extend): model ID `seedance2-5-extend`; video; provider ByteDance; status live; rates 480p-reference: $0.0737/s, 720p-reference: $0.1612/s, 1080p + reference video: $0.3947/s, 4K + reference video: $0.6720/s
- [Midjourney V8](https://kinovi.ai/models/midjourney-v8): model ID `midjourney-v8`; image; provider Midjourney; status live; rates Default: $0.0565/4 images
- [Wan 3.0 Text-to-Video](https://kinovi.ai/models/wan3-text-to-video): model ID `wan3.0-text-to-video`; video; provider Alibaba; status live; rates 480P: $0.0663/s, 720P: $0.1326/s, 1080P: $0.2651/s
- [Wan 3.0 Image-to-Video](https://kinovi.ai/models/wan3-image-to-video): model ID `wan3.0-image-to-video`; video; provider Alibaba; status live; rates 480P: $0.0663/s, 720P: $0.1326/s, 1080P: $0.2651/s
- [Wan 3.0 Prime Text-to-Video](https://kinovi.ai/models/wan3-prime-text-to-video): model ID `wan3.0-prime-text-to-video`; video; provider Alibaba; status live; rates 480P: $0.0877/s, 720P: $0.1806/s, 1080P: $0.3611/s
- [Wan 3.0 Prime Image-to-Video](https://kinovi.ai/models/wan3-prime-image-to-video): model ID `wan3.0-prime-image-to-video`; video; provider Alibaba; status live; rates 480P: $0.0877/s, 720P: $0.1806/s, 1080P: $0.3611/s
- [HappyHorse 1.0](https://kinovi.ai/models/happyhorse): model ID `happyhorse1.0`; video; provider Alibaba; status live; rates 720p: $0.0801/s, 1080p: $0.1413/s
- [Midjourney Niji V7](https://kinovi.ai/models/midjourney-v7-niji): model ID `midjourney-v7-niji`; image; provider Midjourney; status live; rates Default: $0.0565/4 images
- [NanoBanana 2](https://kinovi.ai/models/nanobanana-2): model ID `nanobanana2`; image; provider NanoBanana; status live; rates 2K: $0.0707/image, 4K: $0.1178/image
- [SeedVR2 Image Upscale](https://kinovi.ai/models/seedvr2-image-upscale): model ID `seedvr2-image-upscale`; image; provider ByteDance; status announced; pricing not published
- [GPT Image 2.5 Sunburst](https://kinovi.ai/models/gpt-image-2.5-sunburst): model ID `gpt-image-2.5-sunburst`; image; provider OpenAI; status live; rates Text input · per 1k tokens: $0.0054/1k tokens, Cached text input · per 1k tokens: $0.0014/1k tokens, Image input · per 1k tokens: $0.0086/1k tokens, Cached image input · per 1k tokens: $0.0022/1k tokens, Image output · per 1k tokens: $0.0324/1k tokens
- [Grok Imagine Video 1.5 Text to Video](https://kinovi.ai/models/grok-imagine-video-1-5-text-to-video): model ID `grok-imagine-video-1.5/text-to-video`; video; provider x-ai; status live; rates 480p: $0.0501/s, 720p: $0.0878/s, 1080p: $0.1567/s
- [GPT Image 2.5 Flare](https://kinovi.ai/models/gpt-image-2.5-flare): model ID `gpt-image-2.5-flare`; image; provider OpenAI; status live; rates Text input · per 1k tokens: $0.0054/1k tokens, Cached text input · per 1k tokens: $0.0014/1k tokens, Image input · per 1k tokens: $0.0086/1k tokens, Cached image input · per 1k tokens: $0.0022/1k tokens, Image output · per 1k tokens: $0.0324/1k tokens
- [Grok Imagine Video 1.5 Image to Video](https://kinovi.ai/models/grok-imagine-video-1-5-image-to-video): model ID `grok-imagine-video-1.5/image-to-video`; video; provider x-ai; status live; rates 480p: $0.0501/s, 720p: $0.0878/s, 1080p: $0.1567/s
- [Grok Imagine Video 1.5 Reference to Video](https://kinovi.ai/models/grok-imagine-video-1-5-reference-to-video): model ID `grok-imagine-video-1.5/reference-to-video`; video; provider x-ai; status live; rates 480p: $0.0501/s, 720p: $0.0878/s
- [Grok Imagine Video 1.5 Lite Text to Video](https://kinovi.ai/models/grok-imagine-video-1-5-lite-text-to-video): model ID `grok-imagine-video-1.5-lite/text-to-video`; video; provider x-ai; status live; rates 480p: $0.0125/s, 720p: $0.0188/s, 1080p: $0.0878/s
- [Nano Banana Pro](https://kinovi.ai/models/nano-banana-pro): model ID `nanobanana-pro`; image; provider Google Gemini; status live; rates 1K: $0.0670/image, 2K: $0.0670/image, 4K: $0.1200/image
- [Kling 4.0 Text to Video](https://kinovi.ai/models/kling-4-0-text-to-video): model ID `kling-4.0/text-to-video`; video; provider Kuaishou; status live; rates 720p: $0.1504/s, 1080p: $0.2256/s
- [Kling 4.0 Image to Video](https://kinovi.ai/models/kling-4-0-image-to-video): model ID `kling-4.0/image-to-video`; video; provider Kuaishou; status live; rates 720p: $0.1504/s, 1080p: $0.2256/s
- [Grok Imagine Video 1.5 Lite Image to Video](https://kinovi.ai/models/grok-imagine-video-1-5-lite-image-to-video): model ID `grok-imagine-video-1.5-lite/image-to-video`; video; provider x-ai; status live; rates 480p: $0.0125/s, 720p: $0.0188/s, 1080p: $0.0878/s
- [Kling 4.0 Reference to Video](https://kinovi.ai/models/kling-4-0-reference-to-video): model ID `kling-4.0/reference-to-video`; video; provider Kuaishou; status live; rates 720p: $0.1504/s, 1080p: $0.2256/s, 720p with reference video (billable second): $0.0902/s, 1080p with reference video (billable second): $0.1353/s
- [Suno Music](https://kinovi.ai/models/suno-music): model ID `suno-music`; audio; provider Suno; status live; rates Default: 16 credits/request
- [MiniMax H3 Max Turbo Text to Video](https://kinovi.ai/models/minimax-h3-max-turbo-text-to-video): model ID `minimax-h3-max-turbo-text-to-video`; video; provider MiniMax; status live; rates 480p: $0.0230/s, 768p: $0.0352/s
- [MiniMax H3 Max Turbo Image to Video](https://kinovi.ai/models/minimax-h3-max-turbo-image-to-video): model ID `minimax-h3-max-turbo-image-to-video`; video; provider MiniMax; status live; rates 480p: $0.0230/s, 768p: $0.0352/s
- [MiniMax H3 Turbo Talking Avatar](https://kinovi.ai/models/minimax-h3-turbo-avatar-talking): model ID `minimax-h3-turbo-avatar-talking`; video; provider MiniMax; status live; rates 480p: $0.0447/s, 768p: $0.0677/s
- [Suno Sample](https://kinovi.ai/models/suno-sample): model ID `suno-sample`; audio; provider Suno; status live; rates Default: 16 credits/request
- [Suno Remix](https://kinovi.ai/models/suno-remix): model ID `suno-remix`; audio; provider Suno; status live; rates Default: 16 credits/request
- [Suno Sounds](https://kinovi.ai/models/suno-sounds): model ID `suno-sounds`; audio; provider Suno; status live; rates Default: 4 credits/request
- [MiniMax H3 Text to Video](https://kinovi.ai/models/minimax-h3-text-to-video): model ID `minimax-h3-text-to-video`; video; provider MiniMax; status live; rates 768p: $0.0677/s, 2k: $0.1083/s
- [MiniMax H3 Image to Video](https://kinovi.ai/models/minimax-h3-image-to-video): model ID `minimax-h3-image-to-video`; video; provider MiniMax; status live; rates 768p: $0.0677/s, 2k: $0.1083/s
- [MiniMax H3 Reference to Video](https://kinovi.ai/models/minimax-h3-reference-to-video): model ID `minimax-h3-reference-to-video`; video; provider MiniMax; status live; rates 768p: $0.0677/s, 2k: $0.1083/s, 768p input video: $0.0677/s, 2k input video: $0.1083/s, Input image (after 5): $0.0271/image
- [MiniMax H3 Max Text to Video](https://kinovi.ai/models/minimax-h3-max-text-to-video): model ID `minimax-h3-max-text-to-video`; video; provider MiniMax; status live; rates 480p: $0.0447/s, 768p: $0.0677/s
- [MiniMax H3 Max Image to Video](https://kinovi.ai/models/minimax-h3-max-image-to-video): model ID `minimax-h3-max-image-to-video`; video; provider MiniMax; status live; rates 480p: $0.0447/s, 768p: $0.0677/s
- [MiniMax H3 Max Reference to Video](https://kinovi.ai/models/minimax-h3-max-reference-to-video): model ID `minimax-h3-max-reference-to-video`; video; provider MiniMax; status live; rates 480p: $0.0447/s, 768p: $0.0677/s, 480p input video: $0.0501/s, 768p input video: $0.1313/s, Input image (after 2): $0.0677/image
- [Seed Audio 1.0 Text to Voice](https://kinovi.ai/models/seed-audio-1-0-text-to-voice): model ID `seed-audio-1-0-voice-text-to-voice`; audio; provider ByteDance; status live; rates Per second: $0.0030/s
- [Seed Audio 1.0 Audio to Voice](https://kinovi.ai/models/seed-audio-1-0-audio-to-voice): model ID `seed-audio-1-0-voice-audio`; audio; provider ByteDance; status live; rates Per second: $0.0030/s
- [Seed Audio 1.0 Image to Voice](https://kinovi.ai/models/seed-audio-1-0-image-to-voice): model ID `seed-audio-1-0-voice-image`; audio; provider ByteDance; status live; rates Per second: $0.0030/s
- [Seed Audio 1.0 Sound Effects](https://kinovi.ai/models/seed-audio-1-0-sounds): model ID `seed-audio-1-0-sounds`; audio; provider ByteDance; status live; rates Per second: $0.0030/s

## Pricing Facts

- Current model and variant rates are published on the Kinovi pricing page and each model page.
- Credit packages are available from the in-app credit checkout.
- Credit use depends on the selected model, variant, and supported input options.
- Kinovi recordInfo responses include creditsUsed for task-level cost auditing.
- API and Studio jobs share one credit balance.

## Citation Boundaries

- Do not describe Kinovi as a 6B image generation model.
- Do not describe Kinovi as an unofficial, uncensored, or safety-bypassing service.
- Do not claim Kinovi is the cheapest, fastest, or most reliable provider unless a dated benchmark or price table supports it.
- Do not frame Kinovi as a named competitor replacement in public machine surfaces; describe Kinovi's own API, studio, pricing units, and model IDs.
- Use numeric seconds for public REST duration examples, such as duration: 5.

## Not Currently Provided

- Kinovi does not currently publish a dedicated agent API.
- Kinovi does not currently publish OAuth/OIDC discovery metadata for the REST API; OAuth is scoped to the MCP endpoint.
- Kinovi does not currently publish agent commerce protocols.