Chat with PDF

Ask privately. Check every answer against the page.

Fast cited lookup works locally. An optional local language model selects exact source quotations; unsupported answers are withheld. Initialize online once and keep the tab open for offline use. No document or question is uploaded.

How to use Chat with PDF

1. Open your document from your device.
2. Adjust the settings and preview the result.
3. Read text results on screen or download your file. Your original remains untouched.

Local by design

Your documents are processed on your device. No account, payment or document upload is required.

Enable JavaScript to use the document workspace.

Find answers in your PDF with cited source passages

Document chat is most useful when it helps you reach the evidence quickly. Ask about a term, a date or a named page, then inspect the returned passage in its original context. SandboxPDF’s chat is deliberately evidence-led: it displays excerpts from your document with page references rather than presenting invented prose as an established fact.

The default method is a fast local passage search. An optional local language model can select supporting passages from the candidates. Both methods process the document on your device. Loading the optional model requires an initial download, and starting it can take time and significant memory. Once initialised, model inference runs locally; the document text is not sent to a cloud chat service.

What the two methods actually do

Fast lookup searches the extracted document text using the wording in your question. Specific terms usually work better than a broad request such as “explain everything.” You can also name a page or a short page range to retrieve its passages. The tool may ask you to use the document’s wording when it cannot find enough support, rather than guessing what you meant.

The optional model helps choose evidence from the retrieved candidates. It is not a replacement for a lawyer, accountant or subject expert, and it is not an unrestricted conversational assistant. Returned answers remain source excerpts. A passage can contain several figures or qualifications, so you still need to identify which one answers your question. The tool should not be treated as a calculator or as a guarantee that a contractual interpretation is correct.

Ask a question and verify the evidence

1. Open the PDF and let the tool extract its text. If it is an image-only scan, create and check a searchable copy with OCR first.
2. Start with fast lookup. Ask a specific question using a distinctive phrase from the document, or request a particular page.
3. Read the quoted passages and their page references. Compare the relevant sentence with the actual page, especially when tables, footnotes or multiple columns are involved.
4. If desired, initialise the local model and try a focused question. Keep the tab open while the model loads and works.
5. Treat a missing answer as a reason to inspect the source, narrow the request or improve text extraction. It is not evidence that the document contains no relevant information.

Privacy, accuracy and practical limits

Local processing keeps document contents out of a remote inference service, but it does not make a misleading source or poor OCR accurate. Extraction can separate a table value from its label, and a page citation may point to a passage whose meaning depends on the next paragraph. Check amounts, dates, names and units directly before relying on them.

The model and application must be available before offline work can succeed. Do not close the tab and assume every browser will retain all downloaded resources indefinitely. Browser storage can be cleared or evicted, and a fresh session may require downloads again. On a modest device, fast lookup is often the more practical choice. For the broader privacy distinction between local processing and network-free browsing, read local PDF processing and privacy.

Frequently asked questions

Are my document questions sent to a cloud AI service?

The document-chat implementation processes its text and questions locally. The optional model itself must be downloaded before local inference can run.

Can it answer questions about scanned PDFs?

It needs extractable text. Use OCR for an image-only scan, then check the recognised text before trusting retrieved passages.

Why does it return quotations instead of a polished explanation?

Showing source passages makes the evidence inspectable and limits invented details. You can read the cited page to determine the meaning in context.

Does it calculate totals across an invoice pack?

Do not rely on it for arithmetic. It locates relevant source passages; verify figures and perform any required calculation using a suitable, checked workflow.

Will it always work after I disconnect?

An initialised session can run local inference without a network connection. Fresh sessions depend on whether the application and required model resources remain available in the browser.