Skip to content

A powerful, self-hosted AI chat interface with local RAG (Retrieval-Augmented Generation), file uploads, and support for Google Gemini, OpenAI, Anthropic, and OpenRouter models. Includes session-specific context and persistent history.

License

Notifications You must be signed in to change notification settings

alexandersomosi/freegpt

Repository files navigation

FreeGPT - Private AI Chat with Local RAG

FreeGPT is a modern, privacy-focused AI chat application that you run on your own computer. It allows you to chat with powerful AI models (like Google Gemini, GPT-4, Claude) and even upload your own documents (PDFs, Word files) to ask questions about them.

Képernyőkép 2026-01-11 215417

🚀 Quick Start (Windows Users)

Don't want to install anything?

  1. Go to Releases.
  2. Download the latest FreeGPT.exe.
  3. Double-click to run!
  4. Your browser will open automatically at http://localhost:8000.

(Mac and Linux users: Please follow the "Installation Guide" below to run from source code.)


🌟 Features

  • Chat with Files: Upload documents and the AI will answer based on their content.
  • Private: Your documents stay on your computer.
  • Multi-Model: Use Google, OpenAI, Anthropic, or OpenRouter models.
  • Smart: Automatically reads scanned PDFs (images) using AI vision.
  • Secure: Session isolation keeps your chats private from each other.

📦 Installation Guide (For Developers)

Prerequisites

  • Node.js (v18+)
  • Python (v3.10+)

Step 1: Clone the repository

git clone https://github.com/alexandersomosi/freegpt.git
cd freegpt

Step 2: Set up the Backend

cd backend
python -m venv venv
# Windows: venv\Scripts\activate
# Mac/Linux: source venv/bin/activate
pip install -r requirements.txt
python main.py

Step 3: Set up the Frontend

<img width="2559" height="1305" alt="Képernyőkép 2026-01-11 215814" src="https://github.com/user-attachments/assets/5250924c-260e-4db4-9314-38da9a5990f4" />

# In a new terminal
npm install
npm run dev

🛠️ Build your own EXE

If you want to create your own executable:

  1. Run python build_executable.py in the root directory.
  2. The result will be in the dist/ folder.

📝 License

MIT - Free to use and modify.

Képernyőkép 2026-01-11 220031

About

A powerful, self-hosted AI chat interface with local RAG (Retrieval-Augmented Generation), file uploads, and support for Google Gemini, OpenAI, Anthropic, and OpenRouter models. Includes session-specific context and persistent history.

Topics

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors