Extract Text from Images (OCR)
Pull the text out of screenshots, scans, receipts and photos in 10 languages. Runs offline in your browser, free and unlimited.
Language data downloads once, then recognition runs offline on your machine.
Image
Drop screenshots or photos of text
Scans, receipts, slides, whiteboards. Nothing is uploaded.
Extracted text
How to use the Image to Text
- 1
Choose the language of the text in your image.
- 2
Drop in one or many screenshots, scans or photos.
- 3
Copy the recognised text, or download it as a .txt file.
Frequently asked questions
How accurate is it?
A confidence score is shown for every image. Clean screenshots and typed documents usually score above 90%. Handwriting, low-resolution photos and skewed angles score much lower — if you see a low score, a sharper or straighter photo will help more than anything else.
Does it work offline?
After the language data has downloaded once, yes. The recognition engine is WebAssembly running inside your tab, so you can disconnect from the internet and it keeps working.
Is my data uploaded to a server?
No. Every tool on this site runs entirely inside your browser using JavaScript. Your input never leaves your device and is never sent over the network, which means it is safe to paste production data, API responses and tokens.