Traditional RAG with LanceDB

This example demonstrates how to implement traditional RAG using LanceDB vector database, where knowledge is added to context rather than searched dynamically by the agent.

Code

traditional_rag_lancedb.py

"""
1. Run: `pip install openai lancedb tantivy pypdf sqlalchemy agno` to install the dependencies
2. Run: `python cookbook/rag/03_traditional_rag_lancedb.py` to run the agent
"""

from agno.agent import Agent
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIResponses
from agno.vectordb.lancedb import LanceDb, SearchType

knowledge = Knowledge(
    # Use LanceDB as the vector database and store embeddings in the `recipes` table
    vector_db=LanceDb(
        table_name="recipes",
        uri="tmp/lancedb",
        search_type=SearchType.vector,
        embedder=OpenAIEmbedder(id="text-embedding-3-small"),
    ),
)

knowledge.insert(
    name="Recipes",
    url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf",
)

agent = Agent(
    model=OpenAIResponses(id="gpt-5.2"),
    knowledge=knowledge,
    # Enable RAG by adding references from Knowledge to the user prompt.
    add_knowledge_to_context=True,
    # Set as False because Agents default to `search_knowledge=True`
    search_knowledge=False,
    markdown=True,
)
agent.print_response(
    "How do I make chicken and galangal in coconut milk soup", stream=True
)

Usage

1. Set up your virtual environment

   Mac

   uv venv --python 3.12
   source .venv/bin/activate
   
2. Install dependencies

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

   export OPENAI_API_KEY=your_openai_api_key_here
   
4. Run Agent

   python traditional_rag_lancedb.py