> ## Documentation Index > Fetch the complete documentation index at: https://powower.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # Wan 3.0 Video Generation > Submit an asynchronous Alibaba Cloud Wan 3.0 video generation task. Supports text-to-video, image-to-video (first frame / first-last frame), and reference-to-video. This capability uses the top-level `media` array as multimodal input. Different combinations of `media.type` enable different capabilities: - **Text-to-video**: Only pass `prompt`, no `media`. - **First frame to video**: Pass `first_frame` type media. - **First-last frame to video**: Pass both `first_frame` and `last_frame` type media. - **Reference to video**: Pass `reference_image`, `reference_video`, or `reference_audio` type media. Note: `reference_xx` / `file` / `link` types and `first_frame` / `last_frame` types are mutually exclusive and cannot be mixed in the same request. After submission, query the task status via `GET /v1/videos/{task_id}`. ## OpenAPI ````yaml api-reference/en/zmodelVideo/ali/wan3.0-video-generation.json POST /v1/videos openapi: 3.0.1 info: title: Wan 3.0 Video Generation version: 1.0.0 description: >- Alibaba Cloud Wan 3.0 (wan3.0-video) is an all-in-one reference video generation model that supports text-to-video, image-to-video (first frame / first-last frame), and reference-to-video in a unified interface. Maximum video length is 30 seconds at 30fps. The unified endpoint is `/v1/videos`. license: name: Project License url: https://github.com/QuantumNous/new-api/blob/main/LICENSE servers: - url: https://baze-api.powerbuyin.top description: baze platform API endpoint security: [] tags: - name: Wan 3.0 Video description: Alibaba Cloud Wan 3.0 wan3.0-video video generation capability paths: /v1/videos: post: tags: - Wan 3.0 Video summary: Create Wan 3.0 video generation task description: >- Submit an asynchronous Alibaba Cloud Wan 3.0 video generation task. Supports text-to-video, image-to-video (first frame / first-last frame), and reference-to-video. This capability uses the top-level `media` array as multimodal input. Different combinations of `media.type` enable different capabilities: - **Text-to-video**: Only pass `prompt`, no `media`. - **First frame to video**: Pass `first_frame` type media. - **First-last frame to video**: Pass both `first_frame` and `last_frame` type media. - **Reference to video**: Pass `reference_image`, `reference_video`, or `reference_audio` type media. Note: `reference_xx` / `file` / `link` types and `first_frame` / `last_frame` types are mutually exclusive and cannot be mixed in the same request. After submission, query the task status via `GET /v1/videos/{task_id}`. operationId: aliWan30VideoGenerationCreate requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/Wan30VideoGenerationRequest' examples: file_to_video: summary: File to video value: model: wan3.0-video prompt: >- Generate a presentation video based on the file content, with key information presented in animated form. media: - type: file url: https://example.com/assets/document.pdf size: 1080P ratio: '16:9' seconds: '5' reference_to_video: summary: Reference to video value: model: wan3.0-video prompt: >- Image 1 walks towards the camera, waving and smiling. The background is the garden scene from image 2. media: - type: reference_image url: https://example.com/assets/character.png - type: reference_image url: https://example.com/assets/garden.png size: 1080P ratio: '16:9' seconds: '5' text_to_video: summary: Text to video value: model: wan3.0-video prompt: >- Shot 1: Wide establishing shot, camera starts from a low angle near the ground moving forward, pushing along the grassland direction while tilting upward. size: 1080P ratio: '16:9' seconds: '5' first_frame_to_video: summary: First frame to video value: model: wan3.0-video prompt: >- The person in image 1 slowly turns around and walks towards the camera with a smile. media: - type: first_frame url: https://example.com/assets/opening-frame.png size: 1080P seconds: '5' first_last_frame_to_video: summary: First-last frame to video value: model: wan3.0-video prompt: >- Smoothly transition from the start frame scene to the end frame scene with natural camera movement. media: - type: first_frame url: https://example.com/assets/start-frame.png - type: last_frame url: https://example.com/assets/end-frame.png size: 720P ratio: '16:9' seconds: '5' video_editing: summary: Video editing value: model: wan3.0-video prompt: >- Apply color enhancement and video stabilization to the video. media: - type: reference_video url: https://example.com/assets/source-video.mp4 size: 1080P ratio: '16:9' seconds: '5' video_extension: summary: Video extension value: model: wan3.0-video prompt: >- Continue showing the same scene while maintaining consistent style. media: - type: reference_video url: https://example.com/assets/video-to-extend.mp4 size: 1080P ratio: '16:9' seconds: '10' responses: '200': description: Submission successful, returns a unified video task object. content: application/json: schema: $ref: '#/components/schemas/VideoTaskResponse' '400': description: Invalid request parameters. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: invalid_request: summary: Invalid parameters value: error: message: Invalid request body or field values type: invalid_request_error code: bad_request_body param: '' '401': description: Authentication failed, e.g., missing or invalid token. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: unauthorized: summary: Unauthorized value: error: message: Invalid token type: api_error code: access_denied param: '' '413': description: Request body too large. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: request_too_large: summary: Request body too large value: error: message: Request body too large, please reduce input size type: api_error code: read_request_body_failed param: '' '429': description: Rate limit exceeded or insufficient quota. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: quota_exceeded: summary: Insufficient quota or rate limited value: error: message: Insufficient account quota, please retry later type: api_error code: insufficient_user_quota param: '' '500': description: Internal server error. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: internal_error: summary: Internal error value: error: message: Internal server error, please retry later type: api_error code: gen_relay_info_failed param: '' '502': description: Upstream service returned an invalid response. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: bad_gateway: summary: Bad gateway value: error: message: Upstream service returned an invalid response type: api_error code: bad_response_status_code param: '' '503': description: Upstream service unavailable or no available channels. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' examples: service_unavailable: summary: Service unavailable value: error: message: >- Model temporarily has no available channels, please retry later type: api_error code: get_channel_failed param: '' security: - BearerAuth: [] components: schemas: Wan30VideoGenerationRequest: type: object required: - model properties: model: type: string enum: - wan3.0-video - wan3.0-video-prime description: >- Model name. Options: - `wan3.0-video`: Standard version. - `wan3.0-video-prime`: Speed version with identical capabilities but significantly faster end-to-end speed. prompt: type: string description: >- Text prompt describing the desired video content. Required if `media` is not provided. Supports Chinese and English. Each character counts as one, up to 20000 characters (excess will be truncated). In all-in-one reference mode, use `image 1`, `video 1`, etc. in the prompt to reference media items by their order in the media array. media: type: array description: >- Media asset array supporting images, videos, audio, files, and web links as input. Required if `prompt` is not provided. - Each element is a media object with `type` and `url` fields. - In reference-to-video mode, the order in the array defines the reference order in the prompt. Images and videos are counted separately, so image 1 and video 1 can coexist. - The 1st `reference_image` corresponds to **image 1**, the 2nd to **image 2**, etc. - The 1st `reference_video` corresponds to **video 1**, the 2nd to **video 2**, etc. - The 1st `reference_audio` corresponds to **audio 1**, the 2nd to **audio 2**, etc. items: $ref: '#/components/schemas/Wan30MediaInput' size: type: string enum: - 480P - 720P - 1080P description: >- **Important**: `size` directly affects cost. Please confirm model pricing before calling. Video resolution tier. Default is `1080P`. default: 1080P ratio: type: string enum: - adaptive - '16:9' - '4:3' - '1:1' - '3:4' - '9:16' description: >- Video aspect ratio. Default is `adaptive` (automatically selects an appropriate ratio based on input media and intent). Options: - `adaptive` (default): Adaptive aspect ratio. - `16:9` - `4:3` - `1:1` - `3:4` - `9:16` default: adaptive seconds: type: string description: >- **Important**: `seconds` directly affects cost. Please confirm model pricing before calling. Video duration in seconds. Default is `5`. - Without video input: Integer in range [2, 30]. - With video input: Input video total duration + output video duration must not exceed 30 seconds. - Pass `-1`: Smart duration mode, where the model automatically recommends an appropriate duration based on the prompt, content, and rich media. default: '5' enum: - '-1' - '2' - '3' - '4' - '5' - '6' - '7' - '8' - '9' - '10' - '11' - '12' - '13' - '14' - '15' - '16' - '17' - '18' - '19' - '20' - '21' - '22' - '23' - '24' - '25' - '26' - '27' - '28' - '29' - '30' generate_audio: type: boolean description: |- Whether the output video includes audio. - `true` (default): Output video includes audio. - `false`: Output video has no audio track. Pricing is the same with or without audio. default: true prompt_extend: type: boolean description: >- Whether to enable prompt intelligent rewriting. When enabled, a large model rewrites the input prompt. Significantly improves results for short prompts, but increases processing time. - `true` (default): Enable intelligent rewriting. - `false`: Disable intelligent rewriting. default: true watermark: type: boolean description: >- Whether to add a watermark. The watermark is located in the bottom-right corner with the text "AI Generated". - `false` (default): No watermark. - `true`: Add watermark. default: false seed: type: integer minimum: 0 maximum: 2147483647 description: >- Random seed for reproducibility. Range: [0, 2147483647]. When not specified, a random seed is auto-generated. For better reproducibility, fix the seed value. Note that due to the probabilistic nature of generation, identical results are not guaranteed even with the same seed. VideoTaskResponse: type: object properties: id: type: string description: Unique task identifier. task_id: type: string description: Task ID (alias for id). object: type: string example: video model: type: string description: Model name used. status: type: string enum: - pending - queued - running - succeeded - failed - cancelled description: Task status. progress: type: integer description: Task progress percentage. created_at: type: integer format: int64 description: Creation time as Unix timestamp in seconds. ErrorResponse: type: object properties: error: type: object properties: message: type: string type: type: string description: >- Error type, e.g., `invalid_request_error`, `api_error`, or upstream-compatible error type. code: type: string Wan30MediaInput: type: object required: - type - url properties: type: type: string enum: - first_frame - last_frame - reference_image - reference_video - reference_audio - file - link description: >- Media asset type. Options: - `first_frame`: First frame image. Max 1, strictly used as the video's first frame. - `last_frame`: Last frame image. Max 1, strictly used as the video's last frame. - `reference_image`: Reference image. Max 10. - `reference_video`: Reference video. Max 5 clips, total duration no more than 15 seconds. - `reference_audio`: Reference audio. Max 5 clips, total duration no more than 15 seconds. - `file`: File. Max 1, cannot be used with `link`. - `link`: Web link. Max 1, cannot be used with `file`. **Important**: `reference_xx` / `file` / `link` types and `first_frame` / `last_frame` types are mutually exclusive and cannot be mixed in the same request. url: type: string description: >- Media asset URL. **Image input (type=first_frame / last_frame / reference_image):** Image URL. Formats: JPEG, JPG, PNG (no transparency), BMP, WEBP. Resolution: [240, 8000]px per side, aspect ratio up to 8:1, max file size 20MB. Supported input formats: 1. Public URL: Supports HTTP or HTTPS protocol. Example: `https://xxx/xxx.png`. **Video input (type=reference_video):** Reference video URL. Formats: mp4, mov. Duration: [1, 15]s, resolution: [240, 4096]px per side, aspect ratio up to 8:1, max file size 100MB. Supported input formats: 1. Public URL: Supports HTTP or HTTPS protocol. Example: `https://xxx/xxx.mp4`. **Audio input (type=reference_audio):** Reference audio URL. Formats: wav, mp3. Duration: [1, 15]s, max file size 15MB. Supported input formats: 1. Public URL: Supports HTTP or HTTPS protocol. Example: `https://xxx/xxx.mp3`. **File input (type=file):** File URL. Formats: docx, doc, xlsx, xls, pptx, ppt, pdf, txt, key, pages, numbers, md. Max file size 100MB, max 50 pages. Supported input formats: 1. Public URL: Supports HTTP or HTTPS protocol. Example: `https://xxx/xxx.pdf`. **Web link input (type=link):** Public web page URL. Only supports pages that do not require login. Supported input formats: 1. Public URL: Supports HTTP or HTTPS protocol. Example: `https://xxx/article/xxx`. securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: JWT description: 'Pass `Authorization: Bearer ` in the request header.' ````