PDF-ChatGPT

Paid APILinuxMacSelf-hostedWebWindows

What PDF-ChatGPT is

PDF-ChatGPT is a free, open-source, self-hosted application for asking natural-language questions about PDF files and receiving answers with document and page citations. It uploads and indexes multiple PDFs, showing status, page counts and chunk counts in a document library. Semantic search runs on a persistent FAISS index with a relevance threshold and optional document scope. OpenAI’s Responses API generates answers that can stream token by token, while SQLite stores conversation history for follow-up questions. The browser interface renders Markdown, supports dark and light themes, and adapts to mobile and tablet layouts. PDFs, the FAISS index and conversation history are stored locally, but text is sent to OpenAI for embeddings and chat generation. An OpenAI API key with billing enabled is required, with token-based usage charges. Uploads are limited to 25 MB PDFs; scanned files without extractable text need OCR. The default setup is single-tenant without login, and public exposure is discouraged without added protection.

Who it suits

PDF-ChatGPT suits users who can self-host a web or API application and want cited answers from multiple PDFs. It is relevant to developers comfortable managing OpenAI billing, storage and authentication safeguards.

What it does well

Open-source under the MIT License
Citations identify document and page
Supports multiple PDFs per chat
Conversation history enables follow-ups
Deployable on several hosting platforms

What to know first

OpenAI key and metered billing required
25 MB maximum file size
Scanned PDFs need added OCR
Default setup has no login

Limits

Each upload must be a PDF under the configurable 25 MB limit. Scanned PDFs without an extractable text layer require added OCR, and OpenAI usage incurs metered token charges.

Getting started

Open the project website on GitHub
Run it locally or deploy it to Render, Railway, Fly.io, or Azure App Service
Provide the OpenAI key and settings through environment variables or deployment secrets
Configure persistent storage to retain PDFs, the FAISS index, and SQLite history
Upload PDF files under 25 MB
Ask questions in the web interface or through the API

Verdict

PDF-ChatGPT provides a self-hosted PDF question-answering workflow with semantic retrieval and citations. Its main trade-offs are OpenAI usage costs, file and scan limitations, and the need to secure a default single-tenant deployment.

Questions

Is PDF-ChatGPT free?

The project is free and open source under the MIT License, but OpenAI embedding and chat usage requires billing and is metered per token.

Which platforms can run it?

It supports web, Windows, macOS, Linux, API use, and self-hosted deployment.

Can it handle multiple PDFs?

Yes. It uploads and indexes multiple PDFs and displays status, page counts, and chunk counts in a document library.

Does it provide citations?

Yes. Answers cite the source document and page.

Are scanned PDFs supported?

Scanned PDFs without extractable text are not supported unless OCR is added.

Where is data stored?

Uploaded PDFs and the FAISS index are stored locally, and conversation history is persisted in SQLite; text is sent to OpenAI for embeddings and chat generation.

Facts from PDF-ChatGPT's own pages

Access control: The default setup is single-tenant with no login system, and the maker advises against exposing it publicly without authentication or network restrictions. · github.com, 7 Oct 2026
Answer generation: The app uses OpenAI's Responses API and can stream answers token by token. · github.com, 7 Oct 2026
API and setup help: FastAPI provides autogenerated Swagger, ReDoc, and OpenAPI documentation, and the repository includes a beginner's walkthrough. · github.com, 8 Oct 2026
Chat features: Conversation history is stored in SQLite, prior turns are used for follow-up questions, and answers can stream token by token over Server-Sent Events. · github.com, 8 Oct 2026
Citations: Answers include citations to the source document and page. · github.com, 7 Oct 2026
Conversation features: Conversation history is stored in SQLite and prior turns are used for follow-up questions. · github.com, 7 Oct 2026
Data handling: The maker says PDFs, the FAISS index, and conversation history are stored locally, while text is sent to OpenAI for embeddings and chat generation. · github.com, 7 Oct 2026
Deployment: The README describes local operation and deployment with platforms such as Render, Railway, Fly.io, or Azure App Service, with persistent storage needed to retain data across redeploys. · github.com, 8 Oct 2026
Document library: It supports uploading and indexing multiple PDFs and shows document status, page counts, and chunk counts. · github.com, 7 Oct 2026
File limits: Uploads are restricted to PDF files under a configurable size limit, and scanned PDFs without extractable text are not supported without adding OCR. · github.com, 7 Oct 2026
Integration: The app requires an OpenAI API key and uses OpenAI embeddings; the README says API usage is metered per token and requires billing to be enabled. · github.com, 8 Oct 2026
Integrations: Out of the box, the project uses OpenAI embeddings and the OpenAI Responses API; the maker says adapting it to another LLM provider would require code changes. · github.com, 7 Oct 2026
Interface: The web interface includes Markdown rendering, code syntax highlighting, dark and light themes, and a responsive layout for mobile and tablet. · github.com, 8 Oct 2026
License: The repository states that the project is released under the MIT License. · github.com, 8 Oct 2026
Notable limits: Scanned image PDFs without an extractable text layer are unsupported without adding OCR, and the README lists OCR as a future improvement. · github.com, 8 Oct 2026
PDF library: It supports uploading and indexing multiple PDFs and shows document status, page counts, and chunk counts in a sidebar library. · github.com, 8 Oct 2026
Purpose: PDF-ChatGPT lets users ask natural-language questions about PDFs and receive answers grounded in document content with page citations. · github.com, 8 Oct 2026
RAG stack: The application is built with FastAPI, LangChain, FAISS, and the OpenAI Responses API. · github.com, 8 Oct 2026
Search: Semantic search supports top-k retrieval, a relevance threshold, and optional per-document scoping. · github.com, 8 Oct 2026
Security setup: The README advises configuring the OpenAI key and other settings through environment variables or deployment secrets, and recommends TLS termination through Nginx or Caddy. · github.com, 8 Oct 2026
Storage: Uploaded PDFs and the FAISS index are stored locally, while conversation and message history is persisted in SQLite. · github.com, 8 Oct 2026
Usage costs: Users need an OpenAI API key with billing enabled, and embedding and chat usage is metered per token. · github.com, 7 Oct 2026

Where PDF-ChatGPT stands on Freedom251

AI PDF Assistants#27 of 27