This repo contains an implementation of an Retrieval-Augmented Generation (RAG) chatbot leveraging LlamaIndex, Ollama and Streamlit. The chatbot efficiently retrieves relevant information from documents and generates coherent, contextually appropriate responses. You can test this app in lightning studios via this link, or locally following the steps below.
1. Install ollama and pull models
Start Ollama
ollama serve
Pull the LLM you'd like to use:
ollama pull llama3
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
streamlit run app.py
-
Open localhost:8501 to view your local RAG app.
-
Add PDFs and ask questions. For this case I use this paper about the inadequacy of shap values
-
Example PDF: https://arxiv.org/pdf/2302.08160