Token
A token is the small chunk of text a language model actually reads and writes — often a whole short word, part of a longer word, or a punctuation mark. Models have limits counted in tokens, and many AI services charge by the token.
In one line, for a 12-year-old
Tokens are the puzzle pieces an AI cuts words into before it reads them.
An example
"Unbelievable" might be split into "un", "believ" and "able". In English, 1,000 tokens is roughly 750 words; French usually needs a few more tokens for the same text.
Why it matters to people
Because pricing and limits are counted in tokens, languages that need more tokens can cost more to serve — one reason French and other languages are sometimes treated as second-class.