RAG Workbench

🛠️ Redis RAG Workbench is a development playground for exploring Retrieval-Augmented Generation (RAG) techniques with Redis. Upload a PDF and begin building a RAG app to chat with the document, taking full advantage of Redis features like vector search, semantic caching, LLM memory, and more.

• Integration with Redis for vector storage and caching
• Support for various LLM models and reranking techniques
• Modular architecture for easy extension and customization (soon)

• Python >= 3.11 and Poetry
• Redis Stack

  docker run -d --name redis-stack -p 6379:6379 -p 8001:8001 redis/redis-stack:latest

• OpenAI API key
• Cohere API key (for optional reranking features)

1. Clone the repository:

   git clone https://github.com/redis-developer/redis-rag-workbench.git cd redis-rag-workbench

2. Install the required dependencies with Poetry:

   poetry install --no-root

3. Set up your environment variables by creating a .env file in the project root:

   REDIS_URL=your_redis_url OPENAI_API_KEY=your_openai_api_key COHERE_API_KEY=your_cohere_api_key

To start the application, run:

poetry run uvicorn main:app --reload

This will start the server, and you can access the workbench by navigating to http://localhost:8000 in your web browser.

• main.py: The entry point of the application
• demos/: Contains individual RAG demo implementations
• shared_components/: Reusable utilities and components
• static/: Static assets for the web interface

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

This project is licensed under the MIT License - see the LICENSE file for details.