local-rag

Why we included this project

Local RAG is a self-contained way to get grounded answers over your own documents. It ingests local files, GitHub repositories, and websites, indexes them with Ollama or local Hugging Face embedding models, and keeps everything on your machine or network. The pipeline runs on LlamaIndex and streams responses from local Ollama chat models, so it works well as a reference implementation a team can adapt rather than a hosted service. There are guardrails around uploads, URLs, and repository ingestion, and settings persist in the browser while chat history can be exported, which makes it usable day to day. For anyone wanting to stand up an offline document Q&A system and understand how the pieces fit together, this is a good place to start.