Two related endpoints let you pull structured information out of a PDF rather than just its raw text:
- Document Extract — captures text, tables, or structured information from a PDF, aimed at retrieving key content from documents like invoices, reports, or contracts for reuse elsewhere.
- Field Extraction — extracts a list of fields and their values from an uploaded file (for example invoice numbers, vendor names, or totals). It's designed to work on both digital documents and scans.
Together, these are the building blocks for document-automation use cases like accounts-payable processing or contract data capture.
- Endpoint families: /document-extract/v1 and /field-extraction/v1.