Scanned Document OCR

Most archives are scans with no text layer, which locks them out of search, redaction, and Q&A. Upload a scan, a photo of a page, or a scanned PDF, and two models work through it on your device: EdgeFind-1 locates each piece of text and EdgeRead-1 reads it. Sensitive paperwork never leaves your browser.

EdgeFind-1 + EdgeRead-1 · 96 MBRuns in your browserEnglish printNo Edge Attention servers

Choose a scan or PDF

First 5 pages of a PDF are read. Processed on your device only.

EdgeFind-1 and EdgeRead-1 are built on Apache-2.0 licensed components and prepared for the browser by Edge Attention; notices ship with our repository. Documents are processed on your device and never uploaded.