Chat with any PDF — ask, summarise, understand

Why Chat with a PDF?

PDFs are everywhere — contracts, research papers, financial reports, product manuals, leases and policies — but finding one specific answer often means scrolling through dozens of pages. Chat with PDF turns any document into something you can simply ask. Instead of reading top to bottom, you type a question in plain language and get a direct answer drawn only from your file, with the exact page cited so you can verify it. Students use it to study from lecture notes, professionals to scan contracts for a single clause, and analysts to summarise long reports in seconds. It is the fastest way to understand a document you do not have time to read in full.

How It Works — Smart Retrieval, Not the Whole File

When you upload a PDF, the text is extracted in your browser and only that text is sent to our server — the original file never leaves your device. For each question, a retrieval layer scores every page and sends just the few most relevant ones to the AI, together with a one-time summary of the document that is cached and reused. This means we do not resend the entire PDF for every question, which keeps answers fast and token usage low. The model — Anthropic's Claude Haiku — then answers using only those excerpts and cites the page each fact came from. Because the context is tightly scoped, responses are quick, accurate and grounded in your document rather than guessed.

Private, Free and Limits

Your privacy comes first: the PDF is read locally in the browser, only the extracted text is processed, and nothing is stored permanently — temporary data is deleted within 2 hours. The tool is completely free with no account: you can chat with one PDF of up to 25 pages per day and ask up to 20 questions about it. If your document is longer than 25 pages, use the free Split PDF or Organize PDF tools to keep the section you need, then come back and chat with it. For scanned PDFs with no selectable text, run OCR PDF first so the text can be read.