Create Video

Use `POST /v1/video/create` to submit Veo asynchronous generation tasks through a unified video format.

POSThttps://zx1.deepwl.net/v1/video/create

Request

curl -X POST https://zx1.deepwl.net/v1/video/create \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json" \
  -d '
{
  "prompt": "The cow flew into the sky",
  "model": "veo3.1-fast-components",
  "images": [
    "https://example.com/frame-1.png",
    "https://example.com/frame-2.png"
  ],
  "orientation": "landscape",
  "size": "1280x720",
  "duration": 8,
  "aspect_ratio": "16:9",
  "enable_upsample": false
}'

Response

{
  "id": "video_8b8232a3-847b-40e2-a3a3-912813b7ac60",
  "object": "video",
  "model": "veo3.1-fast-components",
  "status": "queued",
  "progress": 0,
  "created_at": 1768489641,
  "size": "1280x720",
  "detail": {
    "input": {
      "aspect_ratio": "16:9",
      "images": [
        "https://example.com/frame-1.png",
        "https://example.com/frame-2.png"
      ],
      "model": "veo3.1-fast-components",
      "prompt": "The cow flew into the sky"
    }
  },
  "status_update_time": 1768489641
}

The Veo unified video entry point uses POST /v1/video/create, and the request body is JSON. Unlike OpenAI-format Veo video generation, this endpoint uses fields such as images, orientation, and aspect_ratio to represent reference images and aspect ratio.

Reference images are passed as a list of URLs through the images array.
Common model examples include veo3.1-fast-components; use the models actually available in the current channel as the reference.
After a successful submission, a task object containing id, status, and progress is returned. Use Query Task to poll for the result afterward.

Request Headers

Authorizationstring必填

Authentication header. Uses a Bearer token, e.g. Bearer YOUR_API_KEY.

Content-Typestring必填

Request content type, must be application/json.

Request Body

imagesarray必填

An array of image URLs. Pass a list of reference image URLs.

modelstring必填

The model name. For example, veo3.1-fast-components.

orientationstring必填

The frame orientation. portrait means vertical, and landscape means horizontal.

promptstring必填

Prompt. The text describing the video content.

sizestring必填

Output size. Defaults to 720x1280. Available values: 720x1280, 1280x720.

durationinteger必填

Video duration in seconds. Veo defaults to 8 seconds.

aspect_ratiostring必填

Aspect ratio. 16:9 or 9:16.

enable_upsampleboolean

Whether to enable HD upsampling (landscape only).

Request Examples

Response Examples

Response Fields

idstring

Task ID, passed as the id parameter in subsequent queries.

objectstring

Object type, commonly video.

modelstring

The actual model used.

statusstring

Task status. Common values include queued, processing, completed, failed, and cancelled.

progressinteger

Task progress percentage.

created_atinteger

Creation time (Unix timestamp).

sizestring

Output size.

detailobject

Submission details, usually containing the echoed input field.

status_update_timeinteger

Most recent status update time (Unix timestamp).

curl -X POST https://zx1.deepwl.net/v1/video/create \ -H "Authorization: Bearer YOUR_API_KEY" \ -H "Content-Type: application/json" \ -H "Accept: application/json" \ -d ' { "prompt": "The cow flew into the sky", "model": "veo3.1-fast-components", "images": [ "https://example.com/frame-1.png", "https://example.com/frame-2.png" ], "orientation": "landscape", "size": "1280x720", "duration": 8, "aspect_ratio": "16:9", "enable_upsample": false }'