Cosine Similarity
Cosine similarity measures how close two vectors point in the same direction, regardless of their length — the standard way vector search compares a query embedding to product embeddings.
Part of AI Search
Because it ignores magnitude and only measures the angle between vectors, cosine similarity stays stable even when embeddings vary in scale. A score near 1 means the query and product are nearly identical in meaning; a score near 0 means they're unrelated.