Kotaemon is an open-source retrieval-augmented generation (RAG) tool for chatting with your documents, built by Cinnamon AI. It provides a clean, multi-user Gradio interface on top of a full RAG pipeline, so end users can ask questions over PDFs and other documents and get answers with verifiable citations, while developers get a customizable framework they can extend. The project has gathered roughly 25,800 stars on GitHub.
Many RAG demos stop at a notebook; Kotaemon is designed to be deployed and used. It combines hybrid retrieval, full-text and vector search, with reranking to improve answer quality, and it handles multimodal documents containing images and tables, answering questions across all of them. Sources are cited inline and can be previewed in the browser with the relevant passages highlighted, which makes answers easy to verify. Kotaemon runs with local models through Ollama or with API providers, supports multiple users and document collections, and offers advanced options such as GraphRAG-style indexing. The interface targets both everyday users who just want to query their files and developers who need a starting point for a production document-QA system.
Kotaemon is free and open source under the Apache-2.0 license. You host it yourself; the only costs are your infrastructure and any model API usage you choose to enable.
Kotaemon is for teams and individuals who want a ready-to-use, self-hosted chat-with-your-documents application, and for developers who need an open RAG framework with a real user interface rather than a bare library.