
★ 7.2k
March 2026
MIT
Docs & PDF Q&A
What it does
One of the earliest open-source RAG implementations, still deliberately simple: embeddings but no vector database, no indexing layer, no framework — the author's argument being that accuracy matters more than architecture, and that the simple pipeline holds up. Answers cite the page they came from. A hosted demo exists for trying it, which of course means uploading your PDF to someone else's server; run it locally for anything confidential. MIT.
Summary based on the project's own description. The Japanese page has a fuller beginner guide: setup difficulty, what to prepare, and a step-by-step start.
Screenshots & visuals

Before you start
- Setup difficultySome setup needed
- Command lineRequired
- API keyRequired
- Sends data onlineYes
Judged automatically from the project's own metadata. Always confirm on the official page.
pdfGPT FAQ
- Is pdfGPT free?
- pdfGPT itself is free, but it needs an API key for an external AI service. That service charges separately, though some have a free tier.
- Do I need to install anything or write code to use pdfGPT?
- It needs the command line. If you are not comfortable typing commands, expect to get stuck during setup.
- Does pdfGPT work on a phone?
- No. It needs a computer.
- Can I use pdfGPT commercially?
- The licence is MIT, which permits commercial use. Conditions such as keeping the copyright notice still apply, so read the licence text before shipping it inside something.
How it compares
The conditions that decide whether you can actually use it, side by side.
| Tool | Command line | API key | Sends data out | Phone | Commercial use |
|---|---|---|---|---|---|
| pdfGPT (this page) | Required | Required | Yes | No | Yes |
| OCRmyPDF | Required | Not needed | No | No | Yes |
| zerox | Required | Required | Yes | No | Yes |
| paper-qa | Required | Required | Yes | No | Yes |
"Sends data out" means your input is uploaded to someone's server. Pick "No" when the material cannot leave your organisation.