Definition
Plain language
A store of items turned into number lists, so you can look things up by meaning instead of exact words.
As stated in the literature
A system for indexing and querying high-dimensional embeddings via approximate nearest neighbor search, usually with metadata filtering; the standard storage layer behind RAG retrieval.
Also called: vector databases, vector store, vector stores
Why it matters: It is the backbone of most retrieval-augmented generation systems, making it possible to ground LLM answers in large private corpora.
For example, a chatbot can store embeddings of company documents in a vector database and look up the most relevant ones for each user question.
Heard on the show
“A retrieval database, a vector store, a scratchpad.”Episode 192 — A 32B Open Model Matched Frontier Systems By Learning to Take Notes