Personal Introduction
Simple and easy to use, free of charge
Supports importing high-precision models, and even self-training and deployment if needed
Official Introduction
Free, open-source, batch-capable offline OCR software
Compatible with Windows 7 x64 and above
- Free: All code in this project is open-source and completely free.
- Convenient: Ready to use after extraction, runs offline, no network required.
- Efficient: Comes with a highly efficient offline OCR engine and built-in libraries for multiple languages.
- Flexible: Supports multiple invocation methods such as command line and HTTP interface.
- Features: Screenshot OCR / Batch OCR / PDF recognition / QR code / Formula recognition (in testing)
Demo

Download Distribution
You can choose to download via the following methods:
- GitHub Release Umi-OCR v2.1.5 · hiroi-sora/Umi-OCR · GitHub
- Lanzou Cloud https://hiroi-sora.lanzoul.com/s/umi-ocr
- Source Forge Umi-OCR download | SourceForge.net
scoop
scoop bucket add extras
scoop install extras/umi-ocr
Getting Started
The software release package is available as a .7z compressed file or a .7z.exe self-extracting archive. The self-extracting archive can decompress files on computers without any compression software installed.
This software requires no installation. After extraction, just click Umi-OCR.exe to launch the program.
If you encounter any issues, please submit an Issue, and I will try to help you as much as possible.