Graph RAG: LightRAG Integration

LightRAG is a managed knowledge backend that builds a knowledge graph from your documents.

LightRAG is a managed knowledge backend that builds a knowledge graph from your documents. It handles its own ingestion and retrieval, providing graph-based RAG capabilities.

graph_rag.py

"""
Graph RAG: LightRAG Integration
=================================
LightRAG is a managed knowledge backend that builds a knowledge graph
from your documents. It handles its own ingestion and retrieval,
providing graph-based RAG capabilities.

Unlike standard vector-based RAG, LightRAG:
- Extracts entities and relationships from documents
- Builds a knowledge graph for multi-hop reasoning
- Supports graph-traversal queries

Requirements: pip install lightrag-agno
"""

import asyncio

from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIResponses

# ---------------------------------------------------------------------------
# Setup
# ---------------------------------------------------------------------------

try:
    from agno.vectordb.lightrag import LightRag

    knowledge = Knowledge(
        vector_db=LightRag(
            server_url="http://localhost:9621",
        ),
    )

    agent = Agent(
        model=OpenAIResponses(id="gpt-5.2"),
        knowledge=knowledge,
        search_knowledge=True,
        markdown=True,
    )

except ImportError:
    knowledge = None
    agent = None
    print("LightRAG not installed. Run: pip install lightrag-agno")

# ---------------------------------------------------------------------------
# Run Demo
# ---------------------------------------------------------------------------

if __name__ == "__main__":

    async def main():
        if knowledge and agent:
            await knowledge.ainsert(
                url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"
            )

            print("\n" + "=" * 60)
            print("Graph RAG: knowledge graph-based retrieval")
            print("=" * 60 + "\n")

            agent.print_response(
                "What ingredients are commonly shared across Thai recipes?",
                stream=True,
            )

    asyncio.run(main())

Run the Example

1. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   

   uv venv --python 3.12
   .venv\Scripts\activate
   
2. Install dependencies

   uv pip install -U agno openai pypdf
   
3. Export your OpenAI API key

   Mac/Linux

   export OPENAI_API_KEY="your_openai_api_key_here"
   

   $Env:OPENAI_API_KEY="your_openai_api_key_here"
   
4. Run the example

   Save the code above as graph_rag.py, then run:

   python graph_rag.py
   

Full source: cookbook/07_knowledge/04_advanced/03_graph_rag.py