This repository provides a complete Retrieval Augmented Generation pipeline that runs entirely locally using Ollama. Users can upload documents (PDF, TXT, JSON) and query them with an LLM that retrieves relevant context from embedded chunks. It supports query rewriting to improve vague question retrieval, conversation history for multi-turn chats, and an email RAG variant that downloads and indexes emails for querying. Embeddings are generated using models like mxbai-embed-large, stored in a vector database, and served via local Ollama instances running models such as llama3 and mistral.