Loading…
Upload a PDF and ask it questions.
Updated
Turn any PDF into something you can talk to. Upload a document — a report, contract, paper, or manual — and ask questions in plain English to get answers, summaries, and the key points, drawn from the document itself. The text is extracted right in your browser; only that text (not the file) is sent to the AI to answer. It needs a free account: free accounts get 3 AI runs a day shared across every AI tool, and Pro removes the cap.
Drop a PDF, or click to browse
The text is read in your browser, then you can ask questions about it.
Your text is sent to an AI provider to generate the result. We don’t keep it — only a count of your daily runs.
Chatting with a PDF means asking questions in plain English and getting answers drawn from that document rather than from the model's general knowledge. This one is instructed to use only the document's contents and to say when it cannot find something, which is what stops it inventing an answer. The file itself never leaves your browser: the text is extracted on your device, and only that text is sent to be answered, never the PDF. Nothing is stored afterwards. It needs a PDF with selectable text, so a scan or a photographed page will not work until it has been through OCR; the image to text tool handles that. Very long documents are cut at roughly the first 60,000 characters, so answers about the end of a long report may be incomplete. A free account allows three AI runs a day shared across every AI tool.
PDFs with selectable text (exported from Word, Google Docs, most tools). Scanned image-only PDFs won't work unless they've been OCR'd first — try the Image to Text tool for those.
The file itself never leaves your browser — the text is extracted locally, and only that text is sent to the AI to answer your question.
Very long documents are truncated (answers use roughly the first 60,000 characters) to keep responses fast.