v1

latestOpenAPI 3.0.32026-07-241112213.6 KB

Submit Image to Video Generation Task (V2)

Submit an image-to-video generation task using the V2 API (MQ callback mode). Supports multiple input modes: single image, first/last frames, or multiple reference images. Supports multiple AI models with customizable resolution, aspect ratio, duration, and other parameters.

post/v2/common_task/image2video/task/submit

Headers

Topview-Uidstring required

User ID

Authorizationstring required

API Key for authentication

Request body

modelstring required

Model ID, e.g., 'seedance-1.5-pro', 'kling-v3', 'sora-2', etc. (Priority is higher than mode field)

firstFrameFileIdstring

File ID of the first frame image (optional, depends on model input mode)

endFrameFileIdstring

File ID of the end frame image (optional, depends on model input mode)

referenceImageFileIdsstring[]

List of reference image file IDs for multi-image mode. At least 2 images required. Should not include first/end frames (optional)

promptstring

Positive prompt describing the desired video content (optional)

aspectRatiostring

Aspect ratio such as '16:9', '9:16', '1:1', etc. (optional, depends on model support)

resolution480 | 720 | 1080

Video resolution: 480 or 720 or 1080 (optional)

durationinteger

Video duration in seconds. Supported values depend on the model used (optional)

sound'on' | 'off'

Whether to include sound effects: 'on' or 'off' (optional, default: 'off')

generatingCountinteger

Number of videos to generate: 1-4 (optional, default: 1)

boardIdstring

Board ID. When provided, creates a board task record in the system (optional)

noticeUrlstring

Callback URL for task completion notifications. Please refer to the API documentation for details (optional)

Response

Task submitted successfully

codestring

Response status code

messagestring

Response message

Example response

{
  "code": "200",
  "message": "Success"
}