OCRmyPDF is an open-source command-line tool with 35,000+ stars on GitHub that adds a searchable OCR text layer to scanned PDF files.
Scanned documents are just pictures: you cannot search them, copy from them or feed them into a text pipeline. OCRmyPDF fixes that in one step. It renders each page, runs the Tesseract OCR engine, and embeds an invisible, perfectly aligned text layer into a new PDF — the pages look identical, but the document becomes fully searchable and copyable, and stays valid for long-term archiving (PDF/A). It handles deskewing, rotation, image cleanup and multi-language OCR, parallelizes across CPU cores, and ships as a Python tool, a Docker image and distro packages for Linux, Windows, macOS and FreeBSD. First released years ago and still maintained — updated in October 2026 — it is the standard utility behind countless document digitization and RAG ingestion pipelines.
Its longevity is part of the appeal: OCRmyPDF has been the safe, boring answer to searchable PDFs for over a decade, which is exactly what you want in a tool that touches legal, medical and archival documents. It validates its output, refuses to silently corrupt files, and can watch folders or slot into larger pipelines as one dependable step. In the AI era it picked up a second life as the standard pre-processing stage that turns piles of scanned paper into clean, machine-readable text for search indexes, LLM ingestion and document-understanding models.
OCRmyPDF is completely free and open source under the Mozilla Public License 2.0, including for commercial use.
Anyone digitizing paper archives, law firms and offices making scans searchable, and developers building document ingestion for search or AI pipelines.