Redis Vector Database

Use Redis as a vector database for your Knowledge Base.

v2.2.7

You can use Redis as a vector database with Agno.

Setup

uv pip install -U redis redisvl pypdf openai agno

For connecting to a remote Redis instance, pass your Redis connection string to the redis_url parameter and the index name to the index_name parameter of the RedisDB constructor.

For a local docker setup, you can use the following command:

docker run -d --name redis \
  -p 6379:6379 \
  -p 8001:8001 \
  redis/redis-stack:latest

Example

agent_with_knowledge.py

import os

from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.vectordb.redis import RedisDB
from agno.vectordb.search import SearchType

# Configure Redis connection (from environment variables if available, otherwise use local defaults)
REDIS_URL = os.getenv("REDIS_URL", "redis://localhost:6379/0")
INDEX_NAME = os.getenv("REDIS_INDEX", "agno_cookbook_vectors")

# Initialize Redis Vector DB
vector_db = RedisDB(
    index_name=INDEX_NAME,
    redis_url=REDIS_URL,
    search_type=SearchType.vector,  # try SearchType.hybrid for hybrid search
)

# Build a Knowledge base backed by Redis
knowledge = Knowledge(
    name="My Redis Vector Knowledge Base",
    description="This knowledge base uses Redis + RedisVL as the vector store",
    vector_db=vector_db,
)

# Add content (ingestion + chunking + embedding handled by Knowledge)
knowledge.insert(
    name="Recipes",
    url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf",
    metadata={"doc_type": "recipe_book"},
    skip_if_exists=True,
)

# Query with an Agent
agent = Agent(knowledge=knowledge)
agent.print_response("List down the ingredients to make Massaman Gai", markdown=True)

Redis Params

Parameter	Type	Default	Description
index_name	str	Required	Name of the Redis index to store vector data
redis_url	Optional[str]	None	Redis connection URL
redis_client	Optional[Redis]	None	Redis client instance
embedder	Optional[Embedder]	OpenAIEmbedder()	Embedder instance to generate embeddings
search_type	SearchType	SearchType.vector	Type of search to perform (vector, keyword, hybrid)
distance	Distance	Distance.cosine	Distance metric for vector comparisons
vector_score_weight	float	0.7	Weight for vector similarity in hybrid search
reranker	Optional[Reranker]	None	Reranker instance to rerank search results
**redis_kwargs	Any	-	Additional Redis connection parameters