Token

A model doesn’t read letters or whole words: it breaks text into pieces called tokens. In English, a token averages about four characters. Other languages, such as Portuguese, usually take a few more tokens for the same text.

Where it shows up

• Price: APIs charge per million tokens, split into input (what you send) and output (what the model generates). Output usually costs more.
• Limit: the context window is measured in tokens.
• Speed: generation is measured in tokens per second.