ragapp  by ragapp

Agentic RAG app deployable in your own cloud infrastructure

created 1 year ago
4,292 stars

Top 11.6% on sourcepulse

GitHubView on GitHub
Project Summary

This project provides an easy-to-use, self-hostable platform for implementing Agentic Retrieval-Augmented Generation (RAG) in enterprise environments. It aims to offer a configuration experience similar to OpenAI's custom GPTs, but with the flexibility and control of deploying within your own cloud infrastructure via Docker. The primary audience is developers and IT professionals looking to integrate advanced RAG capabilities without the vendor lock-in of cloud-specific solutions.

How It Works

RAGapp is built using LlamaIndex, a popular framework for building LLM applications with data. It leverages Docker for deployment, allowing users to run it in their own cloud environments. The system supports connecting to hosted AI models (OpenAI, Gemini) and local models via Ollama, offering flexibility in model selection. Configuration is managed through an Admin UI, simplifying the setup of RAG pipelines.

Quick Start & Requirements

  • Primary install / run command: docker run -p 8000:8000 ragapp/ragapp
  • Prerequisites: Latest version of Docker. Ollama and Qdrant are required for specific deployment examples.
  • Access Admin UI at http://localhost:8000/admin to configure.
  • Official Docs: Get Started, Endpoints

Highlighted Details

  • Agentic RAG implementation for enterprise use.
  • Configurable via Admin UI, similar to custom GPTs.
  • Supports hosted (OpenAI, Gemini) and local (Ollama) models.
  • Docker-based deployment for self-hosting.

Maintenance & Community

  • Contact: marcusschiesser for questions, feature requests, or bug reports.
  • Development: Requires Poetry for local setup. Frontend builds managed via make build-frontends.

Licensing & Compatibility

  • License: Not explicitly stated in the provided README.

Limitations & Caveats

The project's source code is dynamically retrieved from create-llama, requiring a make build-frontends step before committing changes. Authentication is not included by default and is expected to be handled by an external API Gateway. Authorization features are planned for later versions.

Health Check
Last commit

6 months ago

Responsiveness

1 week

Pull Requests (30d)
0
Issues (30d)
1
Star History
106 stars in the last 90 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of AI Engineering, Designing Machine Learning Systems), Jeremy Howard Jeremy Howard(Cofounder of fast.ai), and
3 more.

cohere-toolkit by cohere-ai

0.2%
3k
RAG toolkit for LLM application development and deployment
created 1 year ago
updated 1 week ago
Feedback? Help us improve.