Get Task Status
Query the execution progress and result of the unified video task interface, and learn the corresponding OpenAI video-compatible query path.
https://zx1.deepwl.net/v1/video/generations/{task_id}curl https://zx1.deepwl.net/v1/video/generations/task-video-abc123 \
-H "Authorization: Bearer YOUR_API_KEY"{
"code": "success",
"message": "",
"data": {
"task_id": "task-video-abc123",
"status": "queued",
"url": "",
"format": "mp4",
"metadata": null,
"error": null
}
}This endpoint is used to query the status of asynchronous tasks submitted through the unified video task entry point. The returned structure is wrapped with code, message, and data.
- The query targets the platform-exposed
task_id, not the upstream real task ID. - A successful response always returns
code = success. - Suitable for use together with
/v1/video/generations. - The same task can also be queried via
/v1/videos/{task_id}in the OpenAIvideoobject format.
Request Headers
Authorizationstring必填Authentication header. Uses a Bearer token, e.g. Bearer YOUR_API_KEY.
Query Parameters
task_idstring必填The public task ID returned by the unified video task entry point.
Request Examples
Response Examples
Response Fields
codestringBusiness status code. Fixed as success on success.
messagestringError or supplementary information. Usually an empty string on success.
dataobjectResponse data container.
task_idstringTask ID.
statusstringTask status. Common values are queued, processing, succeeded, and failed.
urlstringFinal video result URL. Some channels may return a platform proxy URL.
errorobjectError details for failed tasks.
Use Cases
Poll Until Completion
It is recommended to poll every 5 to 10 seconds until status becomes succeeded or failed.
Switch to the OpenAI Query Structure
If the client is already parsing results according to the OpenAI video object, you can directly query:
curl https://zx1.deepwl.net/v1/videos/task-video-abc123 \
-H "Authorization: Bearer YOUR_API_KEY"
Notes
/v1/video/generations/{task_id} and /v1/videos/{task_id} query the same type of video task, but return different structures. The former is a unified task wrapper, while the latter is an OpenAI video object.