Invoice field extraction
Pulls vendor, dates, line items, and totals from scanned invoices in mixed formats. Built on a fine-tuned open model — and when it isn't sure about a field, it flags it for a human instead of guessing.
Input
Scanned PDFs and photos in mixed layouts
Output
Structured fields: vendor, dates, line items, totals
Evaluation
About 96% of fields read correctly on a 400-document test set
Known limit
Handwritten invoices go straight to human review