Vector Store + RAG with Mistral

This project builds a minimal yet powerful vector store in Python, enabling similarity search and retrieval-augmented generation (RAG) with the Mistral LLM via Ollama.

Key features:

• Embedding-based document storage and retrieval
• No need for Langchain or LlamaIndex
• Local setup with simple CLI instructions
• Demonstrated with real research papers

Ideal for learning the core principles behind modern RAG systems.