Token

A token is the unit an AI uses to measure text. Rather than reading letter by letter or word by word, an AI breaks text into these in-between chunks. Some whole words are a single token. Longer or less common words get split into several. A rough rule of thumb for English is that one token is about three quarters of a word, so 1,000 tokens is very roughly 750 words.

You will not usually type or count tokens yourself, so why does the word matter? Because tokens are the currency AI tools use to describe their limits. The size of the context window is given in tokens. If a tool charges for use, the price is often per token. When you see a figure like “128,000 tokens,” that is telling you how much text the AI can handle at once.

The reason AI works in tokens rather than words is partly practical. Tokens give a consistent way to handle any language, along with code, punctuation, and unusual spellings, without needing a fixed dictionary of whole words.

For everyday use, you can keep it simple. Tokens are how AI counts text, and more tokens means more room to work. That is all you need to carry forward.