Vector database
Also known as: vector store, vector search
What is vector database?
A vector database stores text (or images) as numeric embeddings and retrieves items by semantic similarity rather than keyword match. It is the retrieval layer behind most production RAG systems.
Similarity search finds passages that mean the same thing as the query even when they share no words. Most teams do not need a dedicated product — Postgres with pgvector handles millions of chunks comfortably.
Last reviewed by the gAIcko editorial team.