Vector Database

A vector database is purpose-built to handle vector embeddings and enable fast approximate nearest neighbor (ANN) search. Unlike traditional databases that search by exact matches, vector databases find items based on semantic similarity.

Popular vector databases include Pinecone, Weaviate, Qdrant, Milvus, and Chroma. Many traditional databases (PostgreSQL via pgvector, Redis, Elasticsearch) also now support vector search.

Vector databases are essential infrastructure for RAG systems, semantic search, recommendation engines, and AI-powered feature stores. They scale to billions of vectors with millisecond latency using indexing techniques like HNSW (Hierarchical Navigable Small World) graphs.