Marker Result Check
Poll this endpoint to check status of Marker request and retrieve final results
Authorizations
Path Parameters
Response
Successful Response
The status of the request. Should be 'complete' when the request is done.
Estimated overall work completed, from 0 to 100. May pause; not an estimate of time remaining. Null when unavailable. 100 means a successful result is ready.
0 <= x <= 100Signed URL for downloading the completed result JSON when direct result download is required.
Number of seconds until result_url expires.
The format of the output. 'markdown' or 'json'.
The output in chunks format. The top-level key 'blocks' contains a list of blocks from the document with metadata.
The JSON representation of the PDF if the output format is 'json'.
The markdown representation of the PDF if the output format is 'markdown'.
The HTML representation of the PDF if the output format is 'html'.
The filled-in extraction schema. Once confidence scoring finishes, scored fields gain a _score sibling next to their _citations: the confidence that the value is right given its citations. Null fields, fields without citations, and fields whose verification failed have no _score (see extraction_scoring_status).
Average confidence score (0-5) across all extracted fields. Null while scoring is running, or when extraction_scoring_status is 'partial' or 'failed'.
Outcome of background confidence scoring (fast mode). 'complete': every top-level field with a value and citations was scored. 'partial': some fields have no _score sibling and extraction_score_average is null. 'failed': no scores are available. Null while scoring is still running or when scoring does not apply.
complete, partial, failed The extraction mode used for this request: 'turbo', 'fast', 'balanced', or 'accurate'.
Results of document segmentation showing page ranges for each identified segment. Contains segment names, page ranges, and confidence levels (high/medium/low).
The segmentation schema that was applied to this request, echoed back for reference.
A dictionary of the images in the PDF, where the key is the filename for the image, and the value is the base64 encoded image. Images should be stored in the same directory as the PDF.
A dictionary of metadata about the PDF and the conversion process.
Whether the conversion was successful.
If the conversion was not successful, this will contain an error message.
The parse quality score of the output, if available.
The number of pages that were converted.
The total cost of the conversion.
A dictionary of the cost breakdown of this request. Includes the list cost without discounts and final cost to clients after any discounts (e.g. for opting into model training).
The runtime of the conversion.
The ID of the checkpoint that was created for this conversion. This can be used to retrieve the checkpoint later.
A dictionary of the versions of the libraries used in the request.