vector database
A database or extension that stores embedding vectors with their metadata and answers nearest-neighbour queries.
In more detail
Options range from pgvector inside PostgreSQL to dedicated stores such as Qdrant and Chroma and libraries such as FAISS. Choose for the system you already run, its filtering support, scale and operations, after measuring recall on your own data.
Related terms
- pgvector — An open-source PostgreSQL extension that adds a vector column type, distance operators and HNSW and IVF indexes.
- Qdrant — An open-source vector database with HNSW indexing and payload indexes for filtering on metadata.
- Chroma — An open-source embedding database for AI applications, which stores documents, vectors and metadata in collections searched with HNSW.
- FAISS — Meta's open-source library for fast similarity search over dense vectors, with flat, IVF, HNSW and quantised indexes.
- vector index — A data structure that makes nearest-neighbour search over many vectors fast, usually by giving up exactness.
Taught in
- What An Embedding Is — AI Engineering: LLM Apps, RAG & Agents
- Embedding Models — AI Engineering: LLM Apps, RAG & Agents
- Vector Indexes And Databases — AI Engineering: LLM Apps, RAG & Agents
- Retrieval That Finds The Right Thing — AI Engineering: LLM Apps, RAG & Agents
References
- pgvector README: distance operators
- Qdrant documentation: Indexing (HNSW and payload indexes)
- Chroma: Configure collections (HNSW space defaults)