PrivateGPT
RAG & knowledgeAn open-source project for asking questions of your documents fully offline and private.
PrivateGPT answers questions over your documents with no data ever leaving your environment. Ingestion, embedding, retrieval, and generation can all run against local models, so the whole pipeline stays offline.
It was one of the projects that made the private-RAG pattern concrete, and it remains a clear reference implementation of it.
Where it's ideally used
A fit when the absolute requirement is that documents and questions never touch a third-party service.
Where it doesn't fit
Offline-first by design — not the tool when you want the quality of a hosted frontier model in the loop.