PDF AI
A Chrome side panel that reads the document with you — contracts, papers, 100-page reports, even scans. Ask in plain language; check every claim in one click.
Add to Chrome — free · no cardEvery claim carries its page number, quoted exactly when wording matters. No source, no claim — when the file doesn't contain the answer, it says so.
Long PDFs aren't stuffed into the AI. It sees the outline, searches inside by themes, and reads only the sections it needs — you watch every step.
No text layer? Built-in OCR recognizes pages locally in your browser before the chat starts.
Load multiple files at once. Answers name the file and the page they came from.
From the web, a local file, or the Load files button. A chat button appears in the corner of every PDF page.
The text is extracted on your computer. Only the relevant passages are sent to the AI to produce the answer.
Follow the page citation back to the source. Trust, but check — in one click.
Text extraction and OCR happen in your browser. When you ask a question, the relevant document content is sent to our server and the AI service to generate the answer; questions and answers are logged to improve quality. Details in the privacy policy.
Yes. Pages without a text layer are recognized with built-in OCR (English text currently) locally in your browser.
Hundreds of pages is fine. Long documents are searched by meaning rather than sent whole, so answers stay fast and grounded.
Yes — either enable "Allow access to file URLs" for the extension, or simply pick the file with the Load files button.
The assistant understands documents and questions in most major languages. The interface is localized into 16 languages.