Vector database
A database built to store embeddings and quickly find those nearest to a query vector.
Also: vector store, vector index
Finding the closest match among millions of vectors by checking each one is slow, so vector databases use approximate nearest-neighbour indexes, such as HNSW graphs, to find close matches in milliseconds. Products such as Pinecone, Weaviate, Milvus and Qdrant, and vector features in PostgreSQL and other databases, became popular with the growth of retrieval-augmented generation from 2023. They are often combined with keyword search, since exact terms such as product codes are matched better by words than by meaning.