Skip to content
Glossary

Vector Search

Vector search finds content by meaning rather than exact words. Text is converted into high-dimensional embeddings, and a similarity metric such as cosine distance ranks stored vectors by how close they are to the query vector, returning conceptually related passages even when no keywords match.

Synonyms: semantic search, similarity search, nearest-neighbor search, embedding search

Vector search powers semantic retrieval: instead of matching strings, it matches meaning. A query is embedded into the same vector space as the indexed content, and the index returns the nearest vectors by a distance metric. To stay fast at scale, production systems use approximate nearest-neighbor indexes that accept tiny accuracy trade-offs for major latency wins. Vector search is most effective when paired with keyword search in a hybrid retriever, so exact identifiers are not lost to pure semantic matching.

Frequently asked questions

What is an embedding in vector search?
An embedding is a numeric vector that represents the meaning of a piece of text, produced by an embedding model. Texts with similar meaning land close together in the vector space.
What is approximate nearest neighbor (ANN) search?
ANN search trades a small amount of accuracy for large speed gains, using index structures so similarity lookups stay fast as the number of stored vectors grows into the millions.