Embedding

An embedding is a list of numbers that represents the meaning of a piece of text, an image or another item, positioned so that similar things end up numerically close together. "Cancel my subscription" and "how do I stop being billed" land near each other even though they share almost no words.

This is what makes semantic search possible, and semantic search is what makes retrieval-augmented generation work. It is also used for clustering, deduplication and recommendation — anywhere the question is "what else is like this".

Embeddings are produced by a model and stored in a vector database. They are specific to the model that made them: change the embedding model and every stored vector has to be regenerated, which is a migration worth planning for rather than discovering.