Tesseract

Open source

Tesseract is an Apache-2.0 OCR engine with a command line program and C/C++ library, recognising more than 100 languages from PNG, JPEG and TIFF images.

tesseract-ocr.github.io
Tesseract homepage screenshot
GitHub stars
77k
Last 30 days
+572(+1%)
Last commit
2 days ago
Repository age
12 years
Version
5.5.4
Licence
Apache-2.0
Self-hosted
Yes

About Tesseract

Tesseract is an optical character recognition engine. The repository ships both libtesseract and a command line program called tesseract. Version 4 added a neural net engine based on LSTM that works on whole lines, while the older engine from version 3 remains available through the legacy engine mode.

It handles Unicode (UTF-8) text and can recognise more than 100 languages without extra setup. Input can be PNG, JPEG or TIFF, and output can be plain text, hOCR (HTML), PDF, invisible-text-only PDF, TSV, ALTO or PAGE. Further languages can be added through training.

Developers can call the C or C++ API from their own programs, and bindings for other languages are listed in the AddOns documentation. The project began at Hewlett-Packard, was open sourced in 2005 and was developed by Google from 2006 until August 2017. Version 5 is the current stable major version.

Key features

  • LSTM neural net recognition engine
  • Legacy engine mode for Tesseract 3 compatibility
  • Unicode support and more than 100 languages
  • Reads PNG, JPEG and TIFF images
  • Outputs text, hOCR, PDF, TSV, ALTO and PAGE
  • Can be trained for additional languages
  • C and C++ API through libtesseract

Good fit for

  • Extracting text from scanned pages
  • Producing searchable PDFs
  • Embedding OCR in applications through the C or C++ API
Built with
C++ tools
Tags
ocr
text-recognition
cli
library
c++
image-processing
pdf

Tesseract: questions and answers

What is Tesseract used for?
Tesseract is an Apache-2.0 OCR engine with a command line program and C/C++ library, recognising more than 100 languages from PNG, JPEG and TIFF images. It is a good fit for extracting text from scanned pages, producing searchable PDFs and embedding OCR in applications through the C or C++ API.
Is Tesseract open source?
Yes. Tesseract is open source under the Apache-2.0 licence. Its source code is on GitHub at tesseract-ocr/tesseract and is written mainly in C++.
Is Tesseract free?
Yes. Tesseract is open source, so the software itself is free to use.
What are some alternatives to Tesseract?
Similar open-source tools in the Developer Tools category include App Store Connect CLI, TypeUI and Neovim. SaaS products in the same category include act101, Angular Material Blocks and Browserwire.
Is Tesseract actively maintained?
Yes. The most recent commit to Tesseract was on 11 November 2026, and the latest release is 5.5.4, published on 28 October 2026. The project has 77k stars on GitHub, with 572 added in the last 30 days.

Open-source alternatives to Tesseract

See all

SaaS alternatives to Tesseract

See all