Vector Databases provide specialized storage and retrieval for high-dimensional vector embeddings — enabling semantic similarity search and Retrieval-Augmented Generation (RAG) workflows for LLM applications. Core capabilities include indexing algorithms optimized for approximate nearest-neighbor search (HNSW, IVF, DiskANN), hybrid search combining vector similarity with keyword filters and structured metadata, namespace and multi-tenant isolation, horizontal scaling of index size and query throughput, real-time upserts and filtering, and increasingly integrated embedding generation and reranking. The category spans dedicated vector-database vendors (Pinecone, Weaviate, Qdrant, Chroma, Milvus, Zilliz Cloud, Turbopuffer) and general-purpose databases with vector-search extensions (PostgreSQL pgvector, MongoDB Atlas Vector Search, Elasticsearch, Redis).