Definition
Plain language
A specialized database that stores numerical fingerprints of items and finds similar ones quickly.
As stated in the literature
A database optimized for high-dimensional vector storage and approximate nearest-neighbor search, commonly used in retrieval pipelines.
Also called: vector databases
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
“And this matters beyond one paper, because every serious LLM application today runs on a bolt-on vector database doing the searching.”Episode 198 — The Model That Knows the Answer and Can't Say It