What is Token (AI)?

A token is the small chunk of text — roughly three-quarters of a word — that AI models read and generate, and that most AI pricing is based on.

AI models do not read words, they read tokens: fragments of text produced by splitting language into common pieces. As a rough rule, one token is about four characters, and 1,000 tokens is about 750 English words. Tokens matter for two very practical reasons. First, cost: API pricing is quoted per million input and output tokens, so a tool that quietly stuffs long instructions and documents into every request costs more than its headline rate suggests. Second, limits: a model's context window is measured in tokens, so a long document plus a long conversation can exceed what the model can hold at once. Non-English text, code, and unusual names use more tokens per word, which is why costs vary between languages. When comparing tools, translate token pricing into something concrete — the price of a typical 1,500-word article, or a month of support replies — instead of trusting a per-token figure that means nothing on its own.

Also known as

  • tokens
  • token pricing
  • tokenization

Related tools

  • ChatGPT — The most versatile AI assistant for writing, research, and brainstorming.
  • Claude — Thoughtful, safety-focused AI assistant with excellent reasoning.
  • Google Gemini — Google's multimodal AI with real-time information access.
  • Jasper — AI copywriter built for marketing teams with brand voice control.
  • Writesonic — Create SEO-optimized blog posts and landing pages at scale.

Related terms

  • Context Window — The context window is the maximum amount of text, measured in tokens, an AI model can consider at one time.
  • Large Language Model (LLM) — An LLM is an AI model trained on huge amounts of text that predicts language well enough to write, summarise, translate, and answer questions.
  • Generative AI — Generative AI is any AI that creates new content — text, images, audio, video, or code — rather than only analysing existing data.
  • API (Application Programming Interface) — An API is a defined way for one piece of software to request data or actions from another, and it is how AI tools plug into your stack.

Browse all 40 glossary terms

Category: AI Concepts