> ## Documentation Index > Fetch the complete documentation index at: https://docs.omniall.ai/llms.txt > Use this file to discover all available pages before exploring further. # H3 Context IR create > MiniMax-H3 **Context IR** enhanced create. **Path:** `POST https://api.omniall.ai/v1/videos/h3_context_ir` Deep multimodal reasoning. Same fields as create; or set `context_ir_enabled: true` on `POST /v1/videos`. **Mutual exclusion:** `first_frame` / `last_frame` (incl. `image`, `images[0]`, `first_frame_url`) cannot mix with `reference_images` / `reference_videos` / `reference_audios` (or content `reference_*`). **Equivalents:** `first_frame_url` ≈ `first_frame` ≈ `image` ≈ `images[0]`. **ratio rules:** - **t2va:** text-only → `ratio` required, not `adaptive` - **i2va:** with first/last frame → `ratio` forced to `adaptive` - **r2va:** with `reference_*` → `ratio` optional (default `adaptive`) ## OpenAPI ````yaml openapi/minimax-h3-video.yaml POST /v1/videos/h3_context_ir openapi: 3.1.0 info: title: Omniall AI — MiniMax-H3 version: 1.0.0 description: >- MiniMax-H3 async video create. **Path:** `POST https://api.omniall.ai/v1/videos` Also `POST /v1/video/generations`. Supports t2va / i2va / r2va; resolution `480P` / `768P` / `1080P` / `2K`; duration 4–30s. **Mutual exclusion:** `first_frame` / `last_frame` (incl. `image`, `images[0]`, `first_frame_url`) cannot mix with `reference_images` / `reference_videos` / `reference_audios` (or content `reference_*`). **Equivalents:** `first_frame_url` ≈ `first_frame` ≈ `image` ≈ `images[0]`. **ratio rules:** - **t2va:** text-only → `ratio` required, not `adaptive` - **i2va:** with first/last frame → `ratio` forced to `adaptive` - **r2va:** with `reference_*` → `ratio` optional (default `adaptive`) Auth: `Authorization: Bearer sk-...` servers: - url: https://api.omniall.ai security: [] paths: /v1/videos/h3_context_ir: post: tags: - MiniMax-H3 summary: H3 Context IR create description: > MiniMax-H3 **Context IR** enhanced create. **Path:** `POST https://api.omniall.ai/v1/videos/h3_context_ir` Deep multimodal reasoning. Same fields as create; or set `context_ir_enabled: true` on `POST /v1/videos`. **Mutual exclusion:** `first_frame` / `last_frame` (incl. `image`, `images[0]`, `first_frame_url`) cannot mix with `reference_images` / `reference_videos` / `reference_audios` (or content `reference_*`). **Equivalents:** `first_frame_url` ≈ `first_frame` ≈ `image` ≈ `images[0]`. **ratio rules:** - **t2va:** text-only → `ratio` required, not `adaptive` - **i2va:** with first/last frame → `ratio` forced to `adaptive` - **r2va:** with `reference_*` → `ratio` optional (default `adaptive`) operationId: minimaxH3ContextIR parameters: - name: Authorization in: header required: true schema: type: string example: Bearer sk-... description: Bearer API key, e.g. `Bearer sk-...`. requestBody: required: true content: application/json: schema: type: object required: - model - prompt description: MiniMax-H3 create-video body. properties: model: type: string description: 'Model id: `MiniMax-H3`.' example: MiniMax-H3 prompt: type: string description: Text prompt (or `content` text). example: A cat running on grass in bright sunlight duration: type: integer minimum: 4 maximum: 30 description: 'Seconds, 4–30. Alias: `seconds`.' example: 5 seconds: type: string ratio: type: string enum: - adaptive - '21:9' - '16:9' - '4:3' - '1:1' - '3:4' - '9:16' description: >- Aspect ratio. t2va required (not adaptive); i2va forced adaptive; r2va optional. example: '16:9' aspect_ratio: type: string resolution: type: string enum: - 480P - 768P - 1080P - 2K description: '`480P` / `768P` / `1080P` / `2K`.' example: 2K image: type: string description: Single first-frame URL (equiv. `first_frame`). images: type: array items: type: string first_frame: type: string last_frame: type: string first_frame_url: type: string last_frame_url: type: string reference_images: type: array items: type: string reference_videos: type: array items: type: string reference_audios: type: array items: type: string context_ir_enabled: type: boolean description: Enable Context IR. Or use `/v1/videos/h3_context_ir`. callback_url: type: string content: type: array items: type: object properties: type: type: string enum: - text - image_url - video_url - audio_url text: type: string role: type: string image_url: type: object properties: url: type: string video_url: type: object properties: url: type: string audio_url: type: object properties: url: type: string url: type: string examples: text_to_video: summary: Text-to-video (t2va) value: model: MiniMax-H3 prompt: A cat running on grass in bright sunlight duration: 5 resolution: 2K ratio: '16:9' image_to_video_first_frame: summary: Image-to-video · first frame (i2va) value: model: MiniMax-H3 prompt: More steam rises; slow camera push-in duration: 5 resolution: 2K first_frame: https://example.com/first.png image_to_video_first_last: summary: Image-to-video · first & last frame (i2va) value: model: MiniMax-H3 prompt: From morning to dusk, clouds drift duration: 8 resolution: 768P first_frame: https://example.com/morning.png last_frame: https://example.com/dusk.png multimodal_native_content: summary: Multimodal reference · native content (r2va) value: model: MiniMax-H3 content: - type: text text: Character speaking, camera follows - type: image_url image_url: url: https://example.com/char.png role: reference_image - type: video_url video_url: url: https://example.com/motion.mp4 role: reference_video - type: audio_url audio_url: url: https://example.com/voice.mp3 role: reference_audio duration: 5 resolution: 2K ratio: adaptive multimodal_flat: summary: Multimodal reference · flat fields (r2va) value: model: MiniMax-H3 prompt: Keep character look; follow video motion; sync to audio duration: 5 resolution: 2K ratio: '16:9' reference_images: - https://example.com/char.png reference_videos: - https://example.com/motion.mp4 reference_audios: - https://example.com/voice.mp3 responses: '200': description: Accepted content: application/json: schema: type: object properties: id: type: string description: Task id. task_id: type: string status: type: string object: type: string '400': description: Bad request '401': description: Unauthorized ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.