This tutorials build a RAG app powered by LlamaIndex to compare Llama 4 and DeeSeek-R1. We have used Opik for evaluation and observability, which is 100% open-source and nicely integrates with alsmot all popular frameworks.

You can quickly test it on your own complex docs here.

Setup

To sync dependencies, run:

uv sync

Environment Variables

You need to set up the following environment variables:

GROQ_API_KEY=...
OPENAI_API_KEY=...

OpenAI API key needed for using o1 and a judge during evaluation

Ensure these variables are configured correctly before running the application use .env.example as reference and create your own .env file.

Run the streamlit app using streamlit run app.py

Get a FREE Data Science eBook 📖 with 150+ essential lessons in Data Science when you subscribe to our newsletter! Stay in the loop with the latest tutorials, insights, and exclusive resources. Subscribe now!

Contribution

Contributions are welcome! Please fork the repository and submit a pull request with your improvements.

README.md

MetaAI's Llama 4 and DeeSeek-R1 compared using RAG

Setup

Environment Variables

📬 Stay Updated with Our Newsletter!

Contribution