What are Vector Databases?

Vector databases enable us to store information as embeddings and search for “results similar” to our input query using cosine similarity or full text search. These results are then provided to the Agent as context so it can respond in a context-aware manner using Retrieval Augmented Generation (RAG).

Here’s how vector databases are used with Agents:

1. Chunk the information

   Break down the knowledge into smaller chunks to ensure our search query returns only relevant results.

2. Load the knowledge base

   Convert the chunks into embedding vectors and store them in a vector database.

3. Search the knowledge base

   When the user sends a message, we convert the input message into an embedding and “search” for nearest neighbors in the vector database.

Many vector databases also support hybrid search, which combines the power of vector similarity search with traditional keyword-based search. This approach can significantly improve the relevance and accuracy of search results, especially for complex queries or when dealing with diverse types of data.

Hybrid search typically works by:

1. Performing a vector similarity search to find semantically similar content.
2. Conducting a keyword-based search to identify exact or close matches.
3. Combining the results using a weighted approach to provide the most relevant information.

This capability allows for more flexible and powerful querying, often yielding better results than either method alone.

⚡ Asynchronous Operations

Several vector databases support asynchronous operations, offering improved performance through non-blocking operations, concurrent processing, reduced latency, and seamless integration with FastAPI and async agents.

When building with Agno, use the aload methods for async knowledge base loading in production environments.

Supported Vector Databases

The following VectorDb are currently supported:

Azure Cosmos DB MongoDB vCore

Azure Cosmos DB with MongoDB vCore vector search

Cassandra

Apache Cassandra vector database

Chroma

Open-source embedding database

ClickHouse

Fast analytical database with vector search

Couchbase

NoSQL database with vector search*

LanceDB

Fast, local vector database*

LangChain

LangChain vector store integration

LightRAG

Lightweight RAG framework

LlamaIndex

LlamaIndex vector store integration

Milvus

Open-source vector database

MongoDB

MongoDB Atlas vector search

PgVector

PostgreSQL with vector extension*

Pinecone

Managed vector database service*

Qdrant

High-performance vector database

Redis

Redis with vector search capabilities

SingleStore

Real-time analytics database

SurrealDB

SurrealDB vector database

Upstash

Serverless Redis with vector search

Weaviate

Open-source vector database

*hybrid search supported

Each of these databases has its own strengths and features, including varying levels of support for hybrid search and async operations. Be sure to check the specific documentation for each to understand how to best leverage their capabilities in your projects.

Popular Choices by Use Case

Development & Testing

LanceDB - Fast, local, no setup required

Production at Scale

PgVector - Reliable, scalable, full SQL support

Managed Service

Pinecone - Fully managed, no operations overhead

High Performance

Qdrant - Optimized for speed and advanced features

Guides

Getting Started

Build your first knowledge base with a vector database

Embeddings

Learn about creating vector representations of your content

Search & Retrieval

Understand how vector search works with your data

Performance Tips

Optimize your vector database for speed and scale