18 lines (13 loc) · 620 Bytes

AI Study Assistant

A local, AI-powered study assistant that uses Retrieval-Augmented Generation (RAG) to answer questions based on personal study materials. Built with Flask, LangChain, ChromaDB, and Groq (Llama 3).

Features

RAG Architecture: Retrieves relevant information from documents.
Local Vector Store: Uses ChromaDB to store document embeddings.
LLM: Llama 3 (via Groq API).
Multi-Format Support: Ingests PDF, DOCX, and TXT files.

Tech Stack

Python, Flask
Llama 3.1 8B (via Groq API)
LangChain
Chroma (Local)
Embeddings- sentence-transformers/all-MiniLM-L6-v2 (Local)