umbertogriffo/rag-chatbot
A chatbot that uses retrieval-augmented generation to answer questions based on contextual information from markdown files.
Velocity · 7d
+0.4
★ / day
Trend
→steady
star history
This project implements a RAG system that combines a large language model with a vector database for semantic search. It uses ChromaDB as the vector store and runs LLMs locally via llama.cpp, supporting models like llama3 and qwen3-5. Users can query a collection of markdown documents, and the system retrieves relevant context to generate answers.