R2R
StaleDescription
A production-ready Agentic RAG system with RESTful API, featuring multimodal document ingestion, hybrid search, knowledge graph construction, and agent-driven retrieval-augmented generation workflows.
Key Features
- Production-ready Agentic RAG system with RESTful API for retrieval-augmented generation workflows
- Multimodal document ingestion supporting .txt, .pdf, .json, .png, .mp3 and more formats
- Hybrid search combining semantic and keyword search with reciprocal rank fusion
- Automatic knowledge graph construction with entity and relationship extraction
- Deep Research API for multi-step reasoning across knowledgebase and internet sources
- Complete user and access management with authentication and collection-based organization
Use Cases
💡 Building enterprise knowledge management systems with intelligent document retrieval
💡 Creating AI-powered Q&A systems over internal company documents and data
💡 Implementing deep research workflows that combine private data with web sources
💡 Deploying production RAG pipelines with full authentication and access control
Categories
Quick Start
Install: pip install r2r. Set OPENAI_API_KEY. Run: python -m r2r.serve. Use client = R2RClient(base_url='http://localhost:7272') for API access.