RAGEve
A self-hosted retrieval-augmented generation platform that combines Ollama, Qdrant, Redis and MinIO behind a FastAPI backend and Next.js interface, so document search and chat run entirely on a developer's own infrastructure.
RAGEve is a local-first RAG platform aimed at developers and teams who want retrieval-augmented generation without sending data to an external cloud service. It combines Ollama for local LLM inference and embeddings, Qdrant for hybrid dense-and-sparse vector search, and a FastAPI backend with a Next.js web interface, all running on the user's own machine or network.
Storage uses a Peewee ORM with a 27-table schema over MySQL or SQLite, with Redis and MinIO for caching and object storage. An install script provisions uv, Ollama and the required models, and starts the Docker-based services, while developers can instead run the FastAPI and Next.js processes directly for local development.
Recent work has added PDF parsing with table extraction and structured chunking, an evaluation benchmark, and a set of production fixes covering error handling, rate limiting and file upload limits.