Folio On-Device Document OCR

Read documents on your own computer, not someone else’s server. Folio turns PDFs, scans and photographed pages into clean, editable Markdown. Recognition runs entirely on your machine, so your documents never leave it.

Built for real documents
Contracts, research papers, receipts, book chapters and forms — recognized page by page, with the structure kept intact.

Completely private
Recognition happens locally. Nothing is uploaded, queued on a server or shared with anyone. After setup, Folio works with no internet connection at all.

No page limits
Everything runs on your own hardware, so nothing meters your use. Import a one-page form or a five-hundred-page report.

Tables, formulas and charts
Multi-column layouts, tables, mathematical formulas, charts and stamps are reproduced as structured Markdown instead of one flat line of text.

Review and correct
See the original page beside the recognized text, step through pages, and edit anything that came out wrong. Your corrections are kept.

Never lose progress
Each finished page is saved immediately. Close the app or restart the computer mid-document and Folio resumes from where it stopped.

Export your way
Save results as Markdown, plain text, HTML or JSON, one document at a time or in batches, into any folder you choose.Release Notes

System Requirements
Windows 10 version 1809 or later, 64-bit. The engine and model files need several gigabytes of free disk space. At least 8 GB of RAM is recommended; more memory and a faster processor make recognition quicker.

What languages does it support?
The interface is available in English, Simplified and Traditional Chinese, Japanese, Korean, French, German, Spanish and Italian. Text recognition covers considerably more languages than the interface itself offers.