Query Status

GET /v1/video/generations/{id}

Status Values

The unified retrieve endpoint reports data.status with uppercase task states:

Status	Meaning
NOT_START, SUBMITTED, QUEUED	Queued
IN_PROGRESS	Generating
SUCCESS	Completed
FAILURE	Failed; see data.fail_reason

Official-compatible paths of some models (such as Seedance's /api/v3/contents/generations/tasks/{id}) use lowercase statuses queued / running / succeeded / failed; see each model's page for the mapping.

Response Fields

The unified retrieve endpoint returns the task status, result URL, charged quota, and public model information. Some video providers return token usage after the task completes; when the platform can read that data, the response includes usage.

Field	Description
data.task_id	Task ID
data.status	Task status
data.progress	Task progress
data.result_url	Video URL after completion
data.fail_reason	Failure reason when the task fails
data.quota	Platform quota consumed by this task
data.properties.origin_model_name	Public model name requested by the client
data.usage.prompt_tokens	Input tokens; may be 0 or the whole usage object may be omitted if the provider does not return usage
data.usage.completion_tokens	Output/generated tokens
data.usage.total_tokens	Total tokens

{
  "code": "success",
  "message": "",
  "data": {
    "task_id": "task_xxx",
    "status": "SUCCESS",
    "progress": "100%",
    "result_url": "https://example.com/video.mp4",
    "quota": 5635575,
    "properties": {
      "prompt": "Video generation prompt",
      "origin_model_name": "doubao-seedance-2.0"
    },
    "usage": {
      "prompt_tokens": 0,
      "completion_tokens": 731025,
      "total_tokens": 731025
    }
  }
}