Skip to content
Your files stay on your device

File processing runs in your browser. Documents are not uploaded to our servers. The site may download processing engines and send usage or feedback events without file contents.

How we handle your data

Laboratory · On your device

Image to text · compare engines

Recognize a scan with Tesseract and PaddleOCR on your device. Files stay in your browser.

Engines:TesseractPaddleOCRCompare engines

PNG or JPEG · up to 10 MB and 4 megapixels.

Tesseract

Not run

PaddleOCR v5

Not run

PaddleOCR v6 tiny

Not run

Try an example

Download sample

Select English. Both results should include INVOICE, Total: 148.50 and Thank you. Always compare numbers with the original image.

What to expect

Each recognizer has up to 90 seconds. English, Russian and mixed models are available. OCR may misread letters, numbers and reading order; check important values. This tool processes images, not complete PDF documents.

Text output is limited to 1 MB per recognizer.

The first run may download converter files and take a moment. If it fails, check your connection or try a smaller file. Cancel stops active processing; you can choose another file and retry.

Engine licenses and source code