Status Lifecycle

When a document extraction request is submitted, it progresses through the following statuses:

Status Descriptions

REQUESTED

The request has been accepted and queued for processing. The system has validated the input and assigned a request ID. No processing has begun yet.

IN_PROGRESS

The extraction engine is actively processing the document. Depending on document complexity, this may involve OCR, layout analysis, field extraction, and pipeline-specific logic.

COMPLETED

Processing has finished successfully. The extracted results are available in the response body under the result field. Confidence scores and structured data are included.

FAILED

Processing encountered an error and could not complete. The response will include an error object with a code and descriptive message. Common causes include unsupported formats, corrupted files, or pipeline configuration issues.

REVIEW_PENDING

The extraction output requires human review before being finalized. This occurs when a pipeline is configured with a manual review step, typically for low-confidence results or high-stakes documents.

REVIEWED

A reviewer has inspected the extracted data. The results are ready to proceed to the approval stage.

APPROVAL_PENDING

The reviewed results are awaiting final approval. This is an intermediate state between review completion and the final decision.

APPROVED

The extraction results have been approved and are ready for consumption. This is a terminal success state in pipelines that require human sign-off.

REJECTED

The extraction results were rejected during review or approval. This is a terminal failure state. Depending on pipeline configuration, you may resubmit the document.

ℹ

Note: Polling is recommended at intervals of 2–5 seconds. The REQUESTED and IN_PROGRESS statuses are the active polling states. Avoid aggressive polling to prevent rate limiting.