> ## Documentation Index > Fetch the complete documentation index at: https://docs.ttapi.io/llms.txt > Use this file to discover all available pages before exploring further. # Reference to Video > Create reference-to-video task This is the **Wan 3.0 Video Generation API** reference-to-video endpoint. Generate video by uploading reference materials with types `reference_image`, `reference_video`, or `reference_audio`, combined with prompt. ## OpenAPI ````yaml openapi/en/wan-reference-to-video.json POST /wan/api/v1/services/aigc/video-generation/video-synthesis openapi: 3.1.0 info: title: Wan 3.0 - Reference to Video DOCS version: 1.0.0 description: >- Wan 3.0 Reference to Video API - Generate video by uploading reference materials with types `reference_image`, `reference_video`, or `reference_audio`, combined with prompt. servers: - url: https://api.ttapi.io security: [] paths: /wan/api/v1/services/aigc/video-generation/video-synthesis: post: description: Create reference-to-video task requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ReferenceToVideoRequest' examples: reference_image_example: summary: Reference Image Example description: >- Generate video using reference image, supports multiple media type combinations value: model: wan3.0-video input: prompt: >- A peaceful landscape with gentle breeze blowing through the grass media_list: - type: reference_image url: https://example.com/reference-image.jpg - type: reference_video url: https://example.com/reference-video.mp4 - type: reference_audio url: https://example.com/audio.mp3 - type: link url: https://example.com/article - type: file url: https://example.com/document.pdf parameters: resolution: 720P duration: 5 reference_video_example: summary: Reference Video Example description: Generate new video using reference video value: model: wan3.0-video input: prompt: Keep the same style but change the scene media_list: - type: reference_video url: https://example.com/reference-video.mp4 parameters: resolution: 1080P duration: 10 reference_audio_example: summary: Reference Audio Example description: Generate video using reference audio value: model: wan3.0-video input: prompt: Video that matches the rhythm of the audio media_list: - type: reference_audio url: https://example.com/audio.mp3 parameters: resolution: 720P duration: 5 link_example: summary: Web Link Example description: Generate video using web link value: model: wan3.0-video input: prompt: Generate video based on webpage content media_list: - type: link url: https://example.com/article parameters: resolution: 720P duration: 5 file_example: summary: File Example description: Generate video using uploaded file value: model: wan3.0-video input: prompt: Generate video based on file content media_list: - type: file url: https://example.com/document.pdf parameters: resolution: 720P duration: 5 responses: '200': description: Successful response content: application/json: schema: type: object properties: output: type: object properties: task_status: type: string description: Task status enum: - PENDING - RUNNING - SUCCEEDED - FAILED - CANCELED - UNKNOWN example: PENDING task_id: type: string description: Task ID. Valid for query within 24 hours. example: 0385dc79-5ff8-4d82-bcb6-xxxxxx request_id: type: string description: Unique request identifier. example: 4909100c-7b5a-9f92-bfe5-xxxxxx security: - CustomApiKey: [] components: schemas: ReferenceToVideoRequest: type: object description: Reference to video request parameters properties: model: type: string description: |- Model name. Options: - `wan3.0-video`: Standard version. enum: - wan3.0-video example: wan3.0-video input: type: object description: Input basic information. Either prompt or media is required. properties: prompt: type: string description: >- Text prompt describing the expected video content. Supports Chinese and English, up to 20000 characters. In reference-to-video mode, you can use 'Image 1', 'Video 1', 'Audio 1' etc. to refer to media materials in the corresponding order of the media array. media: type: array description: >- Media materials array. In reference-to-video mode, use type: 'reference_image', 'reference_video', 'reference_audio', or 'link' to upload reference materials. **Important**: `reference_xx/file/link` types are mutually exclusive and cannot be mixed in the same request. items: type: object properties: type: type: string description: >- Media material type. Reference-to-video uses the following types: - `reference_image`: Reference image. Up to 10 images. - `reference_video`: Reference video. Up to 5 segments, total duration no more than 15 seconds. - `reference_audio`: Reference audio. Up to 5 segments, total duration no more than 15 seconds. - `link`: Web link. Up to 1 link, cannot be used with file simultaneously. - `file`: File. Maximum 1, cannot be used with link simultaneously. enum: - reference_image - reference_video - reference_audio - link example: reference_image url: type: string description: Media material URL or Base64 encoded data. required: - type - url parameters: type: object description: Video processing parameters. properties: resolution: type: string description: >- Resolution level for generated video. Default is 1080P. Options: 1080P, 720P, 480P default: 1080P enum: - 1080P - 720P - 480P example: 1080P ratio: type: string description: >- Aspect ratio for generated video. Options: adaptive (default), 16:9, 4:3, 1:1, 3:4, 9:16 default: adaptive enum: - adaptive - '16:9' - '4:3' - '1:1' - '3:4' - '9:16' example: adaptive duration: type: integer description: >- Duration of generated video in seconds. Integer range [2, 30], or -1 for smart duration mode. default: 5 example: 10 audio: type: boolean description: >- Whether output video includes audio. true: includes sound (default), false: no audio track. default: true example: true seed: type: integer description: >- Random seed for reproducing generation results. Range: [0, 2147483647]. minimum: 0 maximum: 2147483647 example: 12345 prompt_extend: type: boolean description: >- Whether to enable intelligent prompt rewriting. true: enabled (default), false: disabled. default: true example: true watermark: type: boolean description: >- Whether to add watermark. false: no watermark (default), true: add watermark. default: false example: false required: - model - input securitySchemes: CustomApiKey: type: apiKey in: header name: TT-API-KEY description: >- Please visit TTAPI Dashboard to get your API key. ````