Skip to content

What is Vector database? Definition and explanations

A vector database stores embeddings and quickly retrieves the nearest ones to a given vector, using approximate indexes such as HNSW. It is the search engine behind RAG systems. It can be a dedicated product — Pinecone, Qdrant, Weaviate — or an extension of an existing database, such as pgvector for PostgreSQL, often sufficient at SME scale.

In one sentence

A vector database stores embeddings and quickly retrieves those nearest to a query.

Related terms

Sources and references

Glossary