Vector Database
A vector database stores content as embeddings and is optimized to quickly find the items whose vectors are most similar to a query, powering semantic retrieval at scale.
A vector database is built to search by meaning, fast. It stores embeddings for huge numbers of passages and is engineered to find, in milliseconds, the ones nearest to a query's vector — using approximate nearest neighbor techniques. It's the infrastructure that makes vector search practical at web scale.
You won't touch the database directly as a content creator, but it explains the shape of the game: your passages live in this index as points in meaning-space, and retrieval pulls the closest ones. The cleaner and more focused each passage, the more precisely it's placed — the extractability pillar, viewed from the index side. It's also the same technology many teams use to build their own RAG-powered tools.
Example. A documentation site might load all its pages as embeddings into a vector database so its support assistant can instantly retrieve the most relevant passage for any user question — the same mechanism public answer engines use.
Relevant pillar
Related terms
- EmbeddingsEmbeddings are numerical representations of text that capture its meaning, letting AI systems find passages that are semantically related to a query even when they share no exact keywords.
- Vector SearchVector search is a retrieval method that finds passages by meaning rather than keywords, comparing the numeric embedding of a query against the embeddings of indexed content to surface the closest matches.
- Approximate Nearest Neighbor (ANN)Approximate nearest neighbor is a family of algorithms that quickly find the embeddings most similar to a query without checking every item, making large-scale semantic search fast enough to be practical.