← Back to the A-Z glossary
AI glossary · P1

Token

A token is a small chunk of text the AI counts to measure what it reads, writes, and bills.

Technical basics 4 min read Updated 2026-05-22
— Definition

Token, in plain words

A token is a small chunk of text the AI counts to measure what it reads, writes, and bills.

Give concrete reference points so you can keep cost and length under control.

A concrete example

A short email uses very little. An 80-page PDF sent to an advanced model can add up to tens of thousands of tokens.

Why it matters

The token explains three things at once: cost, maximum length, and response speed.

You'll see it in the pricing tables of AI APIs, in context limits, and in usage counters.

Don't mix it up with

Context window: The context window is how much text a model can take into account in a single conversation or request.

TTFT: TTFT measures the time between sending your request and the first token showing up.

Common mistakes

  • Confusing the number of words with the exact number of tokens.
  • Forgetting that the output often costs more than the input.
  • Running a loop over big documents with no limit.

Quick checklist

  • First I check whether the word names a concept, a tool, a risk, or a metric.
  • I tie it to a concrete case: A short email uses very little. An 80-page PDF sent to an advanced model can add up to tens of thousands of tokens.
  • I keep the main trap in mind: Confusing the number of words with the exact number of tokens.

Quick questions

What is Token in AI?

A token is a small chunk of text the AI counts to measure what it reads, writes, and bills.

Where will I run into Token?

You'll see it in the pricing tables of AI APIs, in context limits, and in usage counters.

Which word should I read next?

Start with Context window, TTFT, TPS.

Want to keep going in order?

Head back to the full glossary, search a word, then open only the pages that deserve more than a short definition.

Open the AI glossary