Tesseract OCR

Package

Open-source OCR engine you self-host that reads 100+ languages from images

Price: Free, open source
Access: None, runs locally

About

Tesseract is the Apache-2.0 OCR engine and CLI behind many OCR tools. It reads text in 100+ languages from images on your own machine and writes plain text, hOCR, PDF, TSV, ALTO or PAGE. Python and JavaScript use community wrappers; image quality matters, so preprocess scans.

What you can do with it

Extract text from scanned images on your own server, with no cloud API
Turn scans into searchable PDF or hOCR output with word positions
OCR documents in 100+ languages and scripts from the command line

Get started

Install it, e.g. brew install tesseract or apt install tesseract-ocr
Add language data such as tesseract-ocr-deu for other languages

Example

# Plain text in output.txt
tesseract scan.png output -l eng

# Searchable PDF in output.pdf, English and German
tesseract scan.png output -l eng+deu pdf

Details

Hosting: Self-hosted, Runs locally
Available in: Worldwide
MCP server: None