Embeddings

Numeric vectors representing text, where similar meanings land close together - the basis of semantic search.

An embedding model turns a piece of text into a list of numbers, typically several hundred to a few thousand. The geometry carries meaning: two passages about the same topic sit near each other even with no shared vocabulary.

That is what makes semantic search possible. Embed your documents once, store them in a vector-database, embed the query at request time, and return the nearest neighbours. It is also the retrieval half of rag.

Embeddings are model-specific. Vectors from two different embedding models are not comparable, so changing the model means re-embedding everything.

Where this comes up

Prompts, configs and tutorials in the library that touch this term.