A chunk of text — roughly a word or part of a word — that a language model processes as a single unit.
LLMs don't read text character by character; they break it into tokens first. Understanding tokens matters for cost (usage is often priced per token) and context limits.
