Definition·
Infrastructure

Vector Database

A database specialized in storing and searching embeddings by semantic similarity.

Detailed explanation

Instead of exact keyword search, a vector database finds content closest in meaning to a query using distance metrics (cosine, dot product). It is at the heart of RAG, semantic search and recommendations.

Examples

pgvector (Postgres)
Pinecone, Weaviate, Qdrant, Milvus
Semantic search across a document base

Frequently asked questions

Do I need a dedicated database?

Not always: pgvector on Postgres is enough for most projects; dedicated DBs shine at very large scale.

Related terms

Last updated: 7/15/2026

Talent AI

Turn theory into practice

Post a mission or join the community of top AI, Data and Machine Learning experts.