Submits an OCR request for asynchronous execution. Returns a job ID immediately with HTTP 202. Poll the corresponding GET endpoint with the job ID to retrieve the result.
Bearer token authentication. Use your provider API key or Bifrost authentication token.
Virtual keys (prefixed with sk-bf-) can also be passed here.
Time-to-live in seconds for the job result after completion. Defaults to 3600 (1 hour). After expiry, the job result is automatically cleaned up.
Model in provider/model format
"mistral/mistral-ocr-latest"
Optional unique identifier for the request
Fallback models in provider/model format
Whether to include base64-encoded images in the response
Specific page indices to process (0-based)
x >= 0Maximum number of images to extract per page
x >= 1Minimum image size in pixels to extract
x >= 1Format for extracted tables (e.g., "markdown", "html")
Whether to extract page headers
Whether to extract page footers
Granularity of confidence scores to include in the response
page, block, word, document Format for bounding box annotations. Supports text, json_object, and json_schema modes.
Format for document-level annotations. Supports text, json_object, and json_schema modes.
Custom prompt for document annotation
Job accepted for processing
Response returned when creating or polling an async job
Unique identifier for the async job
The status of an async job
pending, processing, completed, failed When the job was created
When the job result expires and will be cleaned up
When the job completed (successfully or with failure)
HTTP status code of the completed operation
The result of the completed operation (shape depends on the request type)
Error response from Bifrost