PaddleOCR (92.6K stars) beats MinerU (83.5K) and Tesseract (77.5K) as the most-starred open source OCR project on GitHub. Below are all 138 OCR and document-parsing repos with 1,000+ stars in the Text Mining directory, ranked.

Every RAG pipeline eventually hits the same wall: the PDF. Before a language model can answer anything about your contracts, papers or invoices, something has to turn pixels and layout into clean text. The tools for that job are splitting into two camps. In HackerNoon's comparison of OCR and vision LLMs, template-based OCR breaks when layouts shift, while vision LLMs read documents more flexibly but cost more at volume. The author's verdict: "It could mostly be a hybrid approach."

Meanwhile, the vision models are shrinking. Xiaomi-OCR-0 handles text, tables and formulas with 0.8B parameters, and TeleOCR does structured document parsing with 1.2B. This list is the "show me everything" version of that landscape.

How this list was built

The Text Mining directory tracks the 3,141 most-starred open source text mining projects on GitHub (100+ stars, no forks or archived repos). Its OCR & Documents filter tags 321 of them, which together hold about 1.24M stars. We kept repos with 1,000+ stars (161) and removed 23 that are tagged OCR but aren't really OCR or document tools: generic AI course repos, an inference framework, a clipboard syncer, and captcha solvers, which we chose not to promote. That leaves 138. Star counts are from the directory's November 11, 2026 snapshot.

A few patterns stand out in the full category:

• Python owns it. 158 of the 321 OCR repos are Python. Next are C++ and TypeScript (21 each).
• The licenses are mostly permissive. 111 are MIT and 80 are Apache-2.0. But some heavy hitters are copyleft, including PyMuPDF (AGPL-3.0) and paperless-ngx (GPL-3.0), so check before you ship.
• The new wave is "PDF to LLM-ready Markdown." MinerU, opendataloader-pdf, liteparse, zerox, chandra, Dolphin and GLM-OCR all describe their output as data for AI, not just text.
• Translation is a hidden OCR use case. Manga, visual novel, game and subtitle translators make up a surprising share of the top 50.

Quick picks by job

• PDF to Markdown/JSON for RAG: MinerU, opendataloader-pdf, unstructured, liteparse, zerox, markdownify-mcp
• Classic OCR engines: PaddleOCR, tesseract, EasyOCR, RapidOCR, doctr, ocrs
• OCR in the browser: tesseract.js, Paddle.js
• Make scanned PDFs searchable: OCRmyPDF
• Tables, forms and handwriting: chandra, SimpleHTR, pdftabextract
• Math to LaTeX: LaTeX-OCR, Pix2Text, MixTeX-Latex-OCR
• Desktop screen-grab OCR: Umi-OCR, ShareX, Text-Grab, normcap, TRex
• Self-hosted paperless office: paperless-ngx, paperless-gpt, docspell, naps2
• Smallest/fastest: indra (a 1M-parameter model that runs on CPU), chineseocr_lite, TurboOCR

The full list: 138 open source OCR and document-parsing repos, ranked by stars

1. PaddleOCR by PaddlePaddle: 92.6K stars, Python, Apache-2.0. Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR…
2. MinerU by opendatalab: 83.5K stars, Python, other license. Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your…
3. tesseract by tesseract-ocr: 77.5K stars, C++, Apache-2.0. Tesseract Open Source OCR Engine (main repository).
4. Umi-OCR by hiroi-sora: 48.6K stars, Python, MIT. OCR software, free and offline.
5. paperless-ngx by paperless-ngx: 48K stars, Python, GPL-3.0. A community-supported supercharged document management system: scan, index and archive all…
6. ShareX by ShareX: 40.9K stars, C#, GPL-3.0. ShareX is a free and open-source application that enables users to capture or record any area…
7. tesseract.js by naptha: 38.8K stars, JavaScript, Apache-2.0. Pure Javascript OCR for more than 100 Languages.
8. OCRmyPDF by ocrmypdf: 35.3K stars, Python, MPL-2.0. OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched.
9. opendataloader-pdf by opendataloader-project: 30.3K stars, Java, Apache-2.0. PDF Parser for AI-ready data. Automate PDF accessibility. Open-source.
10. EasyOCR by JaidedAI: 30.1K stars, Python, Apache-2.0. Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin…
11. LaTeX-OCR by lukas-blecher: 16.6K stars, Python, MIT. pix2tex: Using a ViT to convert images of equations into LaTeX code.
12. unstructured by Unstructured-IO: 15.7K stars, HTML, Apache-2.0. Convert documents to structured data effortlessly. Unstructured is open-source ETL solution…
13. Skill_Seekers by yusufkaraaslan: 15.3K stars, Python, MIT. Convert documentation websites, GitHub repositories, and PDFs into Claude AI skills with…
14. Easydict by tisfeng: 15.3K stars, Swift, GPL-3.0. macOS dictionary and translation app with offline OCR.
15. LunaTranslator by HIllya51: 14.1K stars, C++, GPL-3.0. Visual novel translator with OCR text capture.
16. liteparse by run-llama: 13K stars, Rust, Apache-2.0. A fast, helpful, and open-source document parser.
17. chandra by datalab-to: 12.7K stars, Python, Apache-2.0. OCR model that handles complex tables, forms, handwriting with full layout.
18. chineseocr_lite by DayBreak-u: 12.4K stars, C++, GPL-2.0. Ultra-light Chinese OCR with vertical-text support.
19. zerox by getomni-ai: 12.3K stars, TypeScript, MIT. OCR & Document Extraction using vision models.
20. PyMuPDF by pymupdf: 11.1K stars, Python, AGPL-3.0. PyMuPDF is a high performance Python library for data extraction, analysis, conversion &…
21. manga-image-translator by zyddnys: 10.6K stars, Python, GPL-3.0. One-click translation of text inside manga and images.
22. Bob by ripperhe: 9.8K stars, Other, no license. macOS translation and OCR app.
23. video-subtitle-extractor by YaoFANGUK: 9.7K stars, Python, Apache-2.0. Extracts hard-coded subtitles from video into SRT, locally.
24. Dolphin by bytedance: 9.1K stars, Python, other license. The official repo for "Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting"…
25. STranslate by STranslate: 8.6K stars, C#, MIT. A ready-to-go translation ocr tool developed with WPF/WPF.
26. RapidOCR by RapidAI: 8.5K stars, Python, Apache-2.0. Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN…
27. omniparse by adithya-s-k: 7.9K stars, Python, GPL-3.0. Ingest, parse, and optimize any data format from documents to multimedia for enhanced…
28. tessdata by tesseract-ocr: 7.7K stars, Other, Apache-2.0. Trained models with fast variant of the "best" LSTM models + legacy models.
29. GLM-OCR by zai-org: 7.6K stars, Python, Apache-2.0. GLM-OCR: Accurate × Fast × Comprehensive.
30. unstract by Zipstack: 7.4K stars, Python, AGPL-3.0. LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows.
31. eSearch by xushengfeng: 7.4K stars, TypeScript, GPL-3.0. Screenshot, offline OCR, search and translate tool.
32. snow-shot by mg-chao: 6.9K stars, C++, other license. Snow Shot is free, open-source screenshot software and a screen recording tool for Windows and…
33. donut by clovaai: 6.9K stars, Python, MIT. Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic…
34. pdf-craft by oomol-lab: 6.6K stars, Python, MIT. PDF craft can convert PDF files into various other formats. This project will focus on…
35. doctr by mindee: 6.4K stars, Python, Apache-2.0. docTR (Document Text Recognition) - a seamless, high-performing & accessible library for…
36. PaddleX by PaddlePaddle: 6.3K stars, Python, Apache-2.0. All-in-One Development Tool based on PaddlePaddle.
37. Parsr by axa-group: 6.2K stars, JavaScript, Apache-2.0. Transforms PDF, Documents and Images into Enriched Structured Data.
38. chineseocr by chineseocr: 6.1K stars, Python, MIT. YOLOv3 + CRNN Chinese OCR.
39. Translumo by ramjke: 5.9K stars, C#, Apache-2.0. Advanced real-time screen translator for games, hardcoded subtitles in videos, static text and…
40. layout-parser by Layout-Parser: 5.8K stars, Python, Apache-2.0. A Unified Toolkit for Deep Learning Based Document Image Analysis.
41. BallonsTranslator by dmMaze: 5.3K stars, Python, GPL-3.0. Deep-learning-assisted comic translation tool.
42. Text-Grab by TheJoeFin: 5.1K stars, C#, MIT. Use OCR in Windows quickly and easily with Text Grab. With optional background process and…
43. modlens by liustack: 4.9K stars, TypeScript, MIT. The first vision plugin for DeepSeek Harness, and the vision bridge for every text-only coding…
44. naps2 by cyanfish: 4.8K stars, C#, other license. Scan documents to PDF and more, as simply as possible.
45. mmocr by open-mmlab: 4.7K stars, Python, Apache-2.0. OpenMMLab Text Detection, Recognition and Understanding Toolbox.
46. SwiftOCR by NMAC427: 4.6K stars, Swift, Apache-2.0. Fast and simple OCR library written in Swift.
47. KillerPDF by SteveTheKiller: 4.4K stars, C#, GPL-3.0. Free and open-source PDF editor for Windows with a built-in PDF 2.0 engine. View, annotate…
48. deep-text-recognition-benchmark by clovaai: 3.9K stars, Jupyter Notebook, Apache-2.0. Text recognition (optical character recognition) with deep learning methods, ICCV 2019.
49. OXR by SoMarkAI: 3.8K stars, Python, other license. An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during…
50. TextRecognitionDataGenerator by Belval: 3.7K stars, Python, MIT. A synthetic data generator for text recognition.
51. AdelaiDet by aim-uofa: 3.5K stars, Python, other license. AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.
52. text-detection-ctpn by eragonruan: 3.4K stars, Python, MIT. Text detection mainly based on ctpn model in tensorflow, id card detect, connectionist text…
53. CRAFT-pytorch by clovaai: 3.4K stars, Python, MIT. Official implementation of Character Region Awareness for Text Detection (CRAFT).
54. TTime by InkTimeRecord: 3.4K stars, TypeScript, other license. Screenshot, OCR and translation app.
55. Pix2Text by breezedeus: 3.3K stars, Jupyter Notebook, MIT. An Open-Source Python3 tool with SMALL models for recognizing layouts, tables, math formulas…
56. deepdoctection by deepdoctection: 3.3K stars, Python, Apache-2.0. A Repo For Document AI.
57. manga-translator-ui by hgmzhn: 3.2K stars, Python, GPL-3.0. Desktop UI for manga OCR and translation.
58. text-extract-api by CatchTheTornado: 3.2K stars, Python, MIT. Document (PDF, Word, PPTX ...) extraction and parse API using state of the art modern OCRs +…
59. gosseract by otiai10: 3.1K stars, Go, MIT. Go package for OCR (Optical Character Recognition), by using Tesseract C++ library.
60. tools-ocr by AnyListen: 3.1K stars, Java, LGPL-3.0. Local desktop OCR tool (Java).
61. EAST by argman: 3.1K stars, C++, GPL-3.0. A tensorflow implementation of EAST text detector.
62. tesseract-ocr-for-php by thiagoalessio: 3K stars, PHP, MIT. A wrapper to work with Tesseract OCR inside PHP.
63. llm_aided_ocr by Dicklesworthstone: 3K stars, Python, other license. Enhances Tesseract OCR output using LLMs (local or API) for error correction, smart chunking…
64. comic-translate by ogkalu2: 3K stars, Python, Apache-2.0. AI comic and manga translator app/browser extension for automatically translating comics…
65. markdownify-mcp by zcaceres: 3K stars, TypeScript, MIT. A Model Context Protocol server for converting almost anything to Markdown.
66. openrecall by openrecall: 3K stars, Python, AGPL-3.0. OpenRecall is a fully open-source, privacy-first alternative to proprietary solutions like…
67. kordoc by chrisryugj: 3K stars, TypeScript, MIT. Korean HWP/PDF/Office/scan to Markdown converter.
68. CHINESE-OCR by xiaofengShi: 3K stars, Python, no license. End-to-end Chinese scene-text detection and recognition with CTPN, CRNN, and CTC (legacy…
69. TrWebOCR by alisen39: 2.9K stars, Python, Apache-2.0. Offline OCR with a web API, built on the tr engine.
70. manga-ocr by kha-white: 2.8K stars, Python, Apache-2.0. Optical character recognition for Japanese text, with the main focus being Japanese manga.
71. paperless-gpt by icereed: 2.8K stars, Go, MIT. Use LLMs and LLM Vision (OCR) to handle paperless-ngx - Document Digitalization powered by AI.
72. normcap by dynobo: 2.8K stars, Python, other license. OCR powered screen-capture tool to capture information instead of images.
73. retain-pdf by wxyhgk: 2.7K stars, Python, MIT. PDF translation that keeps the original layout.
74. docs by sismics: 2.6K stars, JavaScript, GPL-2.0. Lightweight document management system with OCR.
75. open-paperless by zhoubear: 2.6K stars, Python, other license. Scan, index, and archive all of your paper documents (acquired by Mayan EDMS).
76. better-shot by KartikLabhshetwar: 2.4K stars, Swift, other license. Screenshot, screen recording, and video editor for macOS. Native SwiftUI app with capture…
77. ade-cli by landing-ai: 2.4K stars, Python, Apache-2.0. The official CLI for Agentic Document Extraction (ADE) by LandingAI — parse documents and…
78. mPLUG-DocOwl by X-PLUG: 2.4K stars, Python, Apache-2.0. mPLUG-DocOwl: Modularized Multimodal Large Language Model for Document Understanding.
79. docspell by eikek: 2.4K stars, Elm, AGPL-3.0. Assist in organizing your piles of documents, resulting from scanners, e-mails and other…
80. gowall by Achno: 2.3K stars, Go, MIT. A tool to convert a Wallpaper's color scheme / palette, OCR with VLM's Traditional & Hybrid…
81. claude-real-video by HUANGCHIHHUNGLeo: 2.3K stars, Python, MIT. Let Claude (or any LLM) actually watch a video — scene-aware, deduplicated frames +…
82. pdftabextract by WZBSocialScienceCenter: 2.3K stars, Python, Apache-2.0. A set of tools for extracting tables from PDF files helping to do data mining on…
83. SimpleHTR by githubharald: 2.2K stars, Python, MIT. Handwritten Text Recognition (HTR) system implemented with TensorFlow.
84. deepseek-ocr.rs by TimmyOVO: 2.2K stars, Rust, Apache-2.0. Rust multi-backend OCR/VLM engine (DeepSeek-OCR-1/2, PaddleOCR-VL, DotsOCR) with DSQ…
85. tesserocr by sirfz: 2.2K stars, Python, MIT. A Python wrapper for the tesseract-ocr API.
86. RPA by A9T9: 2.1K stars, TypeScript, other license. Browser and desktop automation with an MCP server for AI assistants. Record macros or write…
87. crnn by bgshih: 2.1K stars, Lua, MIT. Convolutional Recurrent Neural Network (CRNN) for image-based sequence recognition.
88. iOS-OCR-Server by riddleling: 2.1K stars, Swift, MIT. An iOS OCR Server Using Apple's Vision Framework.
89. docext by NanoNets: 2.1K stars, Python, Apache-2.0. An on-premises, OCR-free unstructured data extraction, markdown conversion and benchmarking…
90. gImageReader by manisandro: 2K stars, C++, GPL-3.0. A Gtk/Qt front-end to tesseract-ocr.
91. clawPDF by clawsoftware: 2K stars, C#, AGPL-3.0. Open Source Virtual (Network) Printer for Windows that allows you to create PDFs, OCR text…
92. TRex by amebalabs: 2K stars, Swift, MIT. Copy any text on your screen, stop retyping.
93. ocrs by robertknight: 1.9K stars, Rust, Apache-2.0. Rust library and CLI tool for OCR (extracting text from images).
94. OnnxOCR by jingsongliujing: 1.9K stars, Python, Apache-2.0. ONNX-based OCR runtime derived from PaddleOCR.
95. CRNN_Chinese_Characters_Rec by Sierkinhane: 1.9K stars, Python, no license. (CRNN) Chinese Characters Recognition.
96. MORT by kmonkeyhead: 1.8K stars, C#, MIT. Real-time game translator with OCR.
97. AdvancedLiterateMachinery by AlibabaResearch: 1.8K stars, C++, Apache-2.0. A collection of original, innovative ideas and algorithms towards Advanced Literate Machinery.
98. extractous by yobix-ai: 1.8K stars, Rust, Apache-2.0. Fast and efficient unstructured data extraction. Written in Rust with bindings for many…
99. mokuro by kha-white: 1.8K stars, HTML, GPL-3.0. Read Japanese manga inside browser with selectable text.
100. textshot by ianzhao: 1.8K stars, Python, MIT. Python tool for grabbing text via screenshot.
101. tarsier by reworkd: 1.8K stars, Jupyter Notebook, MIT. Vision utilities for web interaction agents.
102. indra by datalab-to: 1.8K stars, Python, Apache-2.0. 1M param OCR model - runs on CPU, 15x faster than tesseract, close to 1B+ param models in…
103. yomitoku by kotaro-kinoshita: 1.7K stars, Python, no license. Japanese document OCR and layout analysis.
104. MixTeX-Latex-OCR by RQLuo: 1.6K stars, Python, AGPL-3.0. MixTeX multimodal LaTeX, ZhEn, and, Table OCR. It performs efficient CPU-based inference in a…
105. kensaku by yjid: 1.6K stars, Rust, MIT. Search inside documents, code, and even screenshots - instantly and locally. A free…
106. ExtractThinker by enoch3712: 1.6K stars, Python, Apache-2.0. ExtractThinker is a Document Intelligence library for LLMs, offering ORM-style interaction for…
107. docstrange by NanoNets: 1.6K stars, Python, MIT. Extract and convert data from any document, images, pdfs, word doc, ppt or URL into multiple…
108. rpaframework by robocorp: 1.6K stars, Python, Apache-2.0. Collection of open-source libraries and tools for Robotic Process Automation (RPA), designed…
109. PaddleOCR-json by hiroi-sora: 1.6K stars, C++, Apache-2.0. PaddleOCR as an offline JSON API binary.
110. documind by DocumindHQ: 1.5K stars, JavaScript, other license. Open-source platform for extracting structured data from documents using AI.
111. Capso by lzhgus: 1.5K stars, Swift, other license. Open-source screenshot and screen recording for macOS. The free, native alternative to…
112. OpenOCR by Topdu: 1.5K stars, Python, Apache-2.0. OpenOCR: An Open-Source Toolkit for General-OCR Research and Applications, integrates a…
113. Scanner by shouzhong: 1.5K stars, Java, no license. Android scanner for barcodes, ID cards and text.
114. keras-ocr by faustomorales: 1.5K stars, Python, MIT. A packaged and flexible version of the CRAFT text detector and Keras CRNN recognition model.
115. text_renderer by Sanster: 1.5K stars, Python, MIT. Generate text images for training deep learning ocr model.
116. MSPaintIDE by MSPaintIDE: 1.4K stars, Java, MIT. Programming in MS Paint (yes, it OCRs your code).
117. TextSnatcher by RajSolai: 1.4K stars, Vala, GPL-3.0. Copy text from images with OCR on Linux.
118. HRConvert2 by zelon88: 1.4K stars, PHP, GPL-3.0. A self-hosted, resource aware file conversion server supporting 490+ formats in 26 languages.
119. tr by myhub: 1.4K stars, Python, no license. Free offline OCR engine.
120. MouseTooltipTranslator by ttop32: 1.4K stars, JavaScript, MIT. Mouseover Translate Any Language At Once - Chrome Extension: PDF Translator, EBOOK, EPUB, OCR…
121. LaTeX_OCR_PRO by LinXueyuanStdio: 1.3K stars, Jupyter Notebook, GPL-3.0. Math formula image to LaTeX recognition.
122. caffe_ocr by senlinuc: 1.3K stars, C++, no license. Caffe-based OCR experiments.
123. typesafe-computer-use by awlevin: 1.3K stars, Python, MIT. Computer use for about $0.0002 a step: OCR the screen, classify the next action with TypeSafe…
124. CTPN by tianzhi0549: 1.3K stars, Jupyter Notebook, other license. Detecting Text in Natural Image with Connectionist Text Proposal Network (ECCV'16).
125. lift by datalab-to: 1.3K stars, Python, Apache-2.0. Extract structured data from documents quickly and accurately.
126. overlay-translator by ciddwd: 1.3K stars, Kotlin, Apache-2.0. On-screen overlay OCR translator for Android.
127. open-semantic-search by opensemanticsearch: 1.2K stars, Shell, GPL-3.0. Open Source research tool to search, browse, analyze and explore large document collections by…
128. PaddleOCR2Pytorch by frotms: 1.2K stars, Python, Apache-2.0. PaddleOCR inference in PyTorch.
129. MLKit by jenly1314: 1.2K stars, Kotlin, Apache-2.0. Android wrapper for Google ML Kit text recognition.
130. TurboOCR by aiptimizer: 1.2K stars, C++, MIT. TurboOCR, >200 img/s OmnidocBench. TensorRT FP16, PP-OCRv6, HTTP + gRPC.
131. Paddle.js by PaddlePaddle: 1.1K stars, JavaScript, Apache-2.0. Paddle.js is a web project for Baidu PaddlePaddle, which is an open source deep learning…
132. kraken by mittagessen: 1.1K stars, Python, Apache-2.0. OCR engine for all the languages.
133. attention-ocr by emedvedev: 1.1K stars, Python, MIT. A Tensorflow model for text recognition (CNN + seq2seq with visual attention) available as a…
134. SnapX by SnapXL: 1.1K stars, C#, GPL-3.0. SnapX is a free, open-source, cross-platform tool that lets you capture or record any area of…
135. OpenFind by aheze: 1.1K stars, Swift, MIT. An app to find text in real life.
136. anymd by SylphxAI: 1.1K stars, Rust, MIT. Any file → clean Markdown for AI agents: PDF, Word, PowerPoint, Excel, EPUB, HTML and web…
137. capcap by realskyrin: 1.1K stars, Swift, MIT. A lightweight, native macOS screenshot tool that lives in your menu bar.
138. DBNet.pytorch by WenmuZhou: 1K stars, Python, Apache-2.0. A pytorch re-implementation of Real-time Scene Text Detection with Differentiable Binarization.

Want the other 183?

The long tail (100 to 999 stars) is where you find niche tools like single-language engines, receipt parsers and research code. Browse all 321 on the OCR & Documents filter, or narrow by language, license and last-updated date.