Poll Rivya API generation jobs by public task ID, read queued, processing, succeeded, and failed states, and consume result URLs.

Use GET /api/v1/generations/{taskId} to check an asynchronous generation job.

Endpoint

GET https://rivya.ai/api/v1/generations/{taskId}

Required header:

Authorization: Bearer rvya_sk_...

curl Example

curl https://rivya.ai/api/v1/generations/task_public_id \
  -H "Authorization: Bearer rvya_sk_..."

Status Values

Public status values:

queued: task accepted, not complete
processing: generation is running
succeeded: result is available
failed: task failed and may include a public error object

Succeeded Response

{
  "id": "task_public_id",
  "status": "succeeded",
  "model": "z-image",
  "reserved_credits": 1,
  "final_credits": 1,
  "created_at": "2026-05-10T00:00:00.000Z",
  "updated_at": "2026-05-10T00:01:00.000Z",
  "result": {
    "primary_url": "https://media.example/result.png",
    "urls": ["https://media.example/result.png"]
  },
  "error": null
}

primary_url is the first result URL for convenience. urls contains the public result URL list when available.

Failed Response

{
  "id": "task_public_id",
  "status": "failed",
  "model": "z-image",
  "reserved_credits": 1,
  "final_credits": 0,
  "created_at": "2026-05-10T00:00:00.000Z",
  "updated_at": "2026-05-10T00:01:00.000Z",
  "result": null,
  "error": {
    "code": "generation_failed",
    "message": "The generation failed."
  }
}

Polling Guidance

Start with a short delay, then increase polling intervals. Avoid tight loops.

Recommended pattern:

wait 2-3 seconds after submission
poll every 5-10 seconds for normal jobs
slow down for long video and audio jobs
stop polling when status is succeeded or failed

Ownership Boundary

The task must belong to the same Rivya account behind the API key. If not found or not owned by the account, the endpoint returns not_found.

Generation Status