OCRmyPDF
Adds searchable text layer to scanned PDFs using Tesseract OCR.
About
OCRmyPDF is a command-line tool that adds a searchable, selectable text layer to scanned PDFs using the Tesseract OCR engine, leaving the original page images visually unchanged. Along the way it can deskew crooked pages, clean images before recognition, generate archival PDF/A output, and optimize images so the result is often smaller than the input. It supports the 100 plus languages Tesseract covers, validates input and output files, spreads work across CPU cores, and copes with documents thousands of pages long, all without needing a GPU. Installation comes via pip, Homebrew, distribution packages, or Docker on Linux, macOS, Windows through WSL, and the BSDs, and the core is released under the MPL-2.0 license. It is a common backbone for document archiving workflows and document management systems such as Paperless-ngx, and for anyone turning boxes of scans into text they can actually search.
Should you use OCRmyPDF?
Pick it when
Pick OCRmyPDF when you have boxes of scanned PDFs that need to become searchable and selectable while looking unchanged, on ordinary CPUs, with deskewing and PDF/A output for archives or Paperless-ngx.
Look elsewhere when
Skip it when you need structured output such as Markdown, tables, or fields; it only adds a text layer. Recognition quality is Tesseract's, so for handwriting or messy layouts look at olmOCR, Chandra, or Surya.
Alternatives to OCRmyPDF
- Tesseract
The raw engine underneath, for when you want plain text from images in your own code rather than finished, optimized searchable PDFs.
- Docling
Turns scans into structured Markdown or JSON for RAG, rather than preserving the original PDF with a hidden text layer.
- Marker
Produces Markdown with tables and equations from scans, but discards the original page appearance and runs best on a GPU.
- olmOCR
Handles handwriting, old scans, and multi-column reading order with a VLM, but needs an NVIDIA GPU and outputs text instead of a searchable PDF.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- OCR & Document Processing
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Beginner (1/5)
- License
- MPL-2.0
- Added
- Apr 3, 2026