What this is good at
Clean printed text — screenshots, documents, receipts, book pages, signs — comes out nearly perfect. Handwriting is hit-or-miss (that's true of every OCR, including the paid ones). For best results: sharp photo, straight-on angle, decent light.
Why in-browser OCR matters
The popular "image to text" sites upload your image to their servers. Think about what people OCR: leases, IDs, pay stubs, medical letters, homework. This page ships the OCR engine to your browser instead — your image never leaves the device, and it works offline after first load.
FAQ
What languages does it read?
English out of the box (more languages coming — tell us which you need). The engine is Tesseract, the same OCR that powers many paid tools.
Can it do PDFs?
Scanned-PDF OCR is on the workbench. For now: screenshot the page and drop it here.
Why is the first run slow?
Your browser downloads the OCR engine (~few MB) once, then caches it. After that it's fast — and fully offline-capable.