Quivr (39,386 Stars) 2026: Your AI Second Brain - RAG App for Files, Links and Notes

Quivr (39,386 stars) is the open-source RAG app that gives you a second brain - upload files and links, then chat with your knowledge. Here is the complete setup.

💡 What You Will Learn

Quivr (39,386 stars) is the open-source RAG app that gives you a second brain - upload files and links, then chat with your knowledge. Here is the complete setup.

## The short answer **StanGirard/quivr** (39,386 stars, Python) is an opinionated RAG (Retrieval-Augmented Generation) app for integrating GenAI into your apps - and a ready-made second brain. You upload documents, PDFs, and URLs, and Quivr embeds them so you can ask questions and get answers grounded in your own knowledge. ## What Quivr does - **Second brain**: chat with your files, notes, and web links - **RAG API**: expose your knowledge base to your own apps via API - **Multiple formats**: PDF, TXT, Markdown, CSV, and web pages - **Self-hosted**: your data stays on your infrastructure - **Modern stack**: built with FastAPI and React, easy to extend ## Run with Docker Compose ```bash git clone https://github.com/StanGirard/quivr.git cd quivr cp .env.example .env # add your LLM/embedding API keys # backend cd backend && pip install -r requirements.txt && alembic upgrade head uvicorn main:app --reload --port 5050 # frontend cd ../frontend && npm install && npm run dev ``` Or use the hosted version (quivr.app) to try it without setup. ## Using your second brain 1. Create a **brain** (a knowledge base). 2. Upload files or add URLs - Quivr chunks and embeds them. 3. Chat: ask questions; answers cite your sources. 4. Share the brain with your team or expose it via API. ## Practical tips - Organize files into multiple brains (work, research, personal). - Use the API to connect Quivr to Slack, Notion, or custom tools. - Configure embedding model once; switching later requires re-indexing. ## FAQ **Is it free?** Yes - Apache-2.0 open source; cloud hosting is paid. **Do I need my own API keys?** Yes - for self-hosting you supply the LLM and embedding keys. **Can it run fully local?** Yes - point it at local models (Ollama) and local embeddings for a private setup.