Data

Vector database

A database built to store embeddings and find the most similar ones fast.

A vector database indexes thousands or millions of embeddings and answers 'what's closest to this?' in milliseconds — something a normal SQL database is bad at. It's the retrieval engine behind RAG: you embed a user's question, and the vector DB hands back the most relevant chunks of your docs. A regular DB finds rows matching exact values; a vector DB finds rows matching meaning. Plenty of teams just turn on pgvector in Postgres before reaching for a dedicated service.