ragapp
Basic Information
RAGapp provides a deployable, enterprise-focused application for building and running agentic Retrieval-Augmented Generation (RAG) systems. The project aims to make Agentic RAG easy to configure and operate in private cloud environments, offering a ready-made Docker image and management interfaces so teams can run RAG instances on their own infrastructure. It is built using LlamaIndex and supports both hosted models (OpenAI, Gemini) and local models via Ollama. The repo includes an Admin UI for configuration, a Chat UI for end-user interaction, and an API with OpenAPI docs. Deployment examples are provided for Docker Compose (including a single deployment with Ollama and Qdrant and a multi-RAGapp setup with a management UI), and Kubernetes manifests are planned. The source includes development instructions using Poetry and make tasks to build frontends and run in a local development environment.