Folio

Document parsing API

Documents in.
Markdown out.

Fourteen formats into clean Markdown, in about five milliseconds. You only pay OCR rates for the pages that are actually scans.

Runs in this tab

Drop a document. Nothing is uploaded.

The converter is compiled to WebAssembly and runs in your browser, so the file never leaves the page. Open your network tab and watch.

Drop a document

or

.doc .docx .docm .ppt .pptx .pptm .ppsx .xls .xlsx .xlsm .xlsb .odt .ods .odp .pdf .rtf .epub .csv

No upload. Open your network tab and watch.

Reads

Word

  • .doc
  • .docx
  • .docm

PowerPoint

  • .ppt
  • .pptx
  • .pptm
  • .ppsx

Excel

  • .xls
  • .xlsx
  • .xlsm
  • .xlsb

OpenDocument

  • .odt
  • .ods
  • .odp

Portable

  • .pdf
  • .rtf
  • .epub
  • .csv

Where the browser stops

Six pages of that contract are scans. You should pay for six.

42 text · 6 OCR

A page either carries its own text or it is a photograph of words. The first kind converts here, instantly, for nothing. The second needs an OCR model on a server, and that is the only part of this with a real cost behind it.

Most parsing APIs notice one scan and bill OCR rates for the whole document. Folio detects the text layer page by page and charges OCR on the pages that need it — usually a tenth of what you were quoted.

Ask for an API key