Vector

A vector is just an ordered list of numbers — e.g. [0.2, -1.5, 0.8, ...]. It sounds boring, but it’s the universal language of AI: models represent words, images, and ideas as vectors.

Operations on vectors are how models “think”:

• Dot product — a quick measure of how similar two vectors are.
• Addition / scaling — shifting meaning in a controllable way.
• Nearest neighbor — finding the closest vector in a big collection (the core of RAG).