↓ Skip to main content

“There is no server component here, and nothing is uploaded. Your images and documents never leave your computer or phone.”

Simon Willisonsimonwillison.net ↗

Tesseract.js plus Mozilla’s PDF.js, glued together in a few minutes with Claude 3 Opus, gives journalists OCR on a single web page. Multi-column PDFs still come out as one column. Journalists handle documents that should never touch a stranger’s server. This tool has no server to touch.