Topic Hub

Best OCR Open Source Projects

This hub collects the best open source projects in OCR and ranks them by both momentum and authority.

Data window: Last 7 days (with 24h tie-breakers for Trending now)

Last updated: Sep 11, 2026

Projects

20

GitHub stars

31K

Hugging Face likes

2.7K

Replicate runs

11.5K

Authority projects

Projects with the strongest long-term signal in OCR, ranked by total stars, likes, or runs on their primary platform.

Unlimited-OCR

GitHub

GitHub
25Kstars

layerize

Extract editable text layers from graphics for editing and recomposing.

Replicate
6Kruns

glm-ocr

Extract data from documents using a compact multimodal OCR model.

Replicate
5.5Kruns

DeepSeek-OCR-2

Extract text from images and documents with advanced OCR capabilities.

GitHub
1.9Kstars

GLM-OCR

Accurate document understanding and OCR capabilities for various industries and use cases.

GitHub
1.9Kstars

baidu/Qianfan-OCR

Automated document processing and analysis model for extracting information from documents.

Hugging Face Model
1.2Klikes

ocrbase

Extract structured data from PDFs using OCR and LLM-powered parsing.

GitHub
700stars

baidu/Unlimited-OCR

Hugging Face Space

Hugging Face Space
430likes

mac-ocr

Extract text from images and PDFs using a macOS command-line tool.

GitHub
406stars

merterbak/DeepSeek-OCR-Demo

Extract text from images and PDFs to markdown format.

Hugging Face Space
394likes

PaddlePaddle/PaddleOCR-VL-1.6

Document parsing and analysis model for automating document processing tasks.

Hugging Face Model
309likes

light-ocr

GitHub

GitHub
274stars

Hidden gems

Smaller projects with unusually strong momentum. We look for lower total metrics plus positive 7-day growth.

baidu/Unlimited-OCR

Hugging Face Space

Hugging Face Space
430likes
367d

MonkeyOCRv2

GitHub

GitHub
187stars
1877d

SimdPaddleOCR

GitHub

GitHub
217stars
1337d

ATH-MaaS/OvisOCR2

Hugging Face Space

Hugging Face Space
87likes
337d

prithivMLmods/GLM-OCR-Demo

Extract text, formulas, and tables from images with high accuracy.

Hugging Face Space
56likes
267d

datalab-to/chandra-ocr-2

Extract text and layout from images and PDFs with high accuracy.

Hugging Face Model
155likes
267d

lightonai/LightOnOCR-2-1B-Demo

Extract text from images and PDFs using AI-powered OCR technology.

Hugging Face Space
83likes
57d

PaddlePaddle/PaddleOCR-VL-1.6

Document parsing and analysis model for automating document processing tasks.

Hugging Face Model
309likes
187d

merterbak/DeepSeek-OCR-Demo

Extract text from images and PDFs to markdown format.

Hugging Face Space
394likes
37d