Harmondale
Definition

Vector database

A vector database stores embeddings and retrieves the nearest matches to a query vector. It is often used for semantic search and retrieval-augmented generation, where the model needs relevant company documents before answering.

Last updated: 25 June 2026

Why it matters

It helps teams separate model choice from the retrieval layer that often determines answer quality.

Signals to watch

  • Embeddings are indexed
  • Nearest-neighbor search is used
  • Documents are retrieved before generation