Retry a failed/cancelled/expired job in-place (same job ID).
The job is reset to queued and a new Cloud Task is dispatched.
job_id.
Retry is supported when the current status is:
failedcancelledexpired if artifact cleanup has not already completederror, removes the previous response, refreshes expires_at, and requeues the job for execution.
If task dispatch is temporarily delayed after the retry is accepted, the API returns 503 with "Retry queued; dispatch is delayed". The retry still succeeded, so poll the same job_id.Successful Response
Core Job object following OpenAI-style specification.
Represents a single asynchronous job that can be polled for status and result retrieval.
/v1/documents/extract, /v1/documents/parse, /v1/documents/split, /v1/documents/classify, /v1/schemas/generate, /v1/edit/agent/fill, /v1/edit/templates/fill, /v1/edit/templates/generate, /v1/evals/extract/process, /v1/evals/split/process, /v1/evals/classify/process, /v1/evals/extract/extract, /v1/evals/extract/split "job"validating, queued, in_progress, completed, failed, cancelled, expired Error details when job fails.
Response stored when job completes successfully.