Local RAG
An offline retrieval-augmented generation tool that ingests local files, GitHub repositories and websites, then answers questions over them using local Ollama models with no third party ever seeing the data.
Local RAG is an open-source tool for retrieval-augmented generation that runs entirely offline with local large language models through Ollama, keeping chat, embeddings and indexed source content on the user's own machine or network rather than sending it to a third party.
It ingests content from several source types, including local files, GitHub repositories and websites, and can use either Ollama or local Hugging Face models for embeddings. Chat responses stream through LlamaIndex as the retrieval pipeline runs, settings persist in the browser, and chat history can be exported. Ingestion includes guardrails around uploads, URLs and repository content.
It is written in Python and licensed under the GNU General Public License version 3.