Umi-OCR

Umi-OCR: free offline OCR for screenshots, images, and PDFs

Umi-OCR is a free, open-source OCR application for users who want text recognition without uploading documents to a cloud service. It runs locally on Windows and Linux, making it useful for private documents, scanned files, screenshots, receipts, manuals, academic papers, and multilingual images.

What it can do

• Screenshot OCR: capture any screen area with a hotkey and turn visible text into editable text.
• Batch image OCR: process many images at once without online usage limits.
• PDF OCR: extract text from scanned PDFs and create searchable double-layer PDF files.
• QR and barcode recognition: decode QR codes plus many common barcode formats from images.
• Math formulas: convert formulas in images into LaTeX-style output where supported.
• Offline privacy: recognition happens on the local machine, so sensitive files do not need to leave the device.

Paddle vs Rapid builds

The community site describes two common packages: Paddle for higher accuracy with a larger model, and Rapid for faster recognition with a smaller footprint. Choose Paddle for complex documents and Rapid for quick everyday OCR.

Screenshot field and visual reference

Who should use it?

Umi-OCR is a strong pick for students, translators, researchers, office workers, and developers who need private OCR on Windows or Linux. It is less suitable if you need a hosted API, collaborative OCR workspace, or enterprise document processing pipeline.

Sources checked

• Official Umi-OCR GitHub repository
• Official releases
• Community information site, not affiliated with the official project