Vision Study RAG

A vision-based RAG (Retrieval-Augmented Generation) system for studying PDFs with charts and images, using the ColPali vision language model.

Features (Coming Soon)

• PDF processing with visual awareness of images, charts, and tables
• No OCR required - works directly with the visual content
• Vector storage using ChromaDB (local, free, open-source)
• Simple and intuitive web interface
• Compatible with Ollama for completely local LLM generation

Important: Python 3.13 Compatibility Notice

The repository is being updated to work with Python 3.13. Some dependencies need adaptation to work correctly with this version.

Quick Start (Simplified Version)

1. Clone this repository

   git clone https://github.com/tofunori/vision-study-rag.git
   cd vision-study-rag

2. Run the simplified setup script

   setup_simplified.bat

3. Run the application

   run_app.bat

4. Open your browser at http://localhost:8501

Development Status

This project is under active development. The current version includes:

• Basic Streamlit interface
• Simplified setup for Python 3.13
• Roadmap for full implementation

Future updates will add:

• Full ColPali integration
• Vector database connectivity
• PDF processing
• LLM integration with Ollama

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.