Quivr
Open-source second-brain platform that lets you chat with your documents and knowledge base using any LLM and a customizable RAG pipeline.
About
Quivr is an open-source retrieval-augmented generation platform that turns your uploaded documents, databases, and connected tools into a searchable, conversational knowledge base. You can upload PDFs, CSVs, text files, and other formats, then ask questions across all of them in a single chat interface powered by your choice of model, including GPT-4, Claude, Llama, or Mistral.
The platform is built to be customizable: you can swap vector stores, add internet search, build custom tools, and integrate it into existing products via its API. A hosted free plan includes 100 message credits and access to three knowledge bases. Paid plans unlock higher limits, and the full source code is available on GitHub for self-hosting at no cost.
The platform is built to be customizable: you can swap vector stores, add internet search, build custom tools, and integrate it into existing products via its API. A hosted free plan includes 100 message credits and access to three knowledge bases. Paid plans unlock higher limits, and the full source code is available on GitHub for self-hosting at no cost.