AI glossary
Embedding
A list of numbers that represents the meaning of a piece of text (or an image), so that similar meanings end up close together.
An embedding model turns text into a vector — often hundreds or thousands of numbers. Texts about similar things produce vectors that are near each other, even if they use different words. That makes embeddings the basis of semantic search, recommendations and RAG.
Example: “How do I reset my password?” and “I forgot my login” have very different words but very similar embeddings.
In practice
- Use the same embedding model for your documents and your questions: vectors from different models cannot be compared.
- Split long documents into meaningful chunks, by section or paragraph, before converting them.
- If you switch embedding model, you will have to recompute all the vectors.
