The OCR API is unable to process documents at this moment. We are investigating the source of the issues.
The performance of the OCR API is back to normal.
The origin of the issue has been identified and a fix has been placed. The OCR API might respond slower while it's processing the backlog.