Read the text out of scanned PDFs with on-device OCR — your document never leaves your device.
No — and it is worth being exact. On the first run your browser downloads the OCR engine (its worker script and WebAssembly core) and the recognition model for your language, roughly 2–15 MB in total, from a public CDN; all of it is cached afterwards. Those are one-way downloads. Your PDF travels the other way only — it is rendered and read inside this tab, so the document itself never leaves your device.
It depends on scan quality — clean 200+ dpi scans of printed text usually come out very well. We show the engine’s own confidence score with every result so you know how much proofreading to expect.
English, Spanish, French, German and Portuguese. Pick the language before running — using the wrong model sharply reduces accuracy.
This tool runs entirely in your browser. Your file is processed on your device and is not uploaded to a server.
Pull all selectable text out of a PDF, with word count, ready to copy or download.
Convert every PDF page into a high-quality image, at the resolution you choose.
Reduce PDF file size with clear before/after numbers — ideal for scans and image-heavy files.
Attempt to fix a corrupt PDF by re-parsing and rebuilding its file structure.