What is Semantic Search?
Search that ranks results by meaning similarity between the query and documents, using embeddings, rather than by matching keywords.
Traditional search engines match and weight the words in your query. Semantic search instead compares what the query means to what each document means, using embedding vectors. A search for "rotate credentials" surfaces the page titled "regenerate API keys" even though they share no significant words.
Semantic search is also naturally cross-lingual with multilingual embedding models, and robust to vague or conversational phrasing — the way real users actually ask.
Its weakness is precision on literal strings, which is why it is usually deployed as one half of a hybrid system rather than alone.
How BeforeQuery uses this
Semantic retrieval over your indexed chunks is the vector half of BeforeQuery's hybrid search, and it is what lets users ask in their own words — or their own language — and still land on the right document, with a citation to verify.
Related terms
Embeddings
Numeric vector representations of text in which semantically similar passages end up close together, enabling search by meaning rather than exact words.
Hybrid Search
A retrieval strategy that combines semantic (vector) search with keyword (full-text) search and merges the results, capturing both meaning and exact matches.
Vector Database
A data store optimized for saving embedding vectors and finding the nearest neighbors of a query vector quickly, at scale.
From definition to deployment
Connect a knowledge source and see cited AI answers on your own content in minutes — free, no credit card required.
Get Started Free