Skip to main content

54. Embeddings and Vector Search

Embedding models convert data into vectors.

Similar items should have similar vectors.

A common similarity measure is cosine similarity:

cos(θ)=A⋅B∥A∥∥B∥cos(\theta)= \frac{A\cdot B} {\|A\|\|B\|}

Vector search retrieves vectors that are close to a query vector.

This forms the foundation of semantic search and RAG.