Skip to content
Bae
Glossary

Token (AI) — definition and meaning

A token is the basic unit of text an AI model processes — roughly a word or word-piece. Models read and generate text token by token, and limits like the context window are measured in tokens.

Last reviewed 2026-05-25

Tokens are how language models chop up text: common words may be a single token, while longer or rarer words split into several. A useful rule of thumb is ~0.75 words per token in English. Tokens matter to AI companions because every limit and cost is measured in them — the context window (how much the model remembers), pricing (most APIs bill per token), and response length. When a companion 'runs out of memory,' it's running out of token budget in the window.

Common questions

About token (ai).

What is token (ai)?

A token is the basic unit of text an AI model processes — roughly a word or word-piece. Models read and generate text token by token, and limits like the context window are measured in tokens.

How is "token (ai)" used in AI companion apps?

Tokens are how language models chop up text: common words may be a single token, while longer or rarer words split into several. A useful rule of thumb is ~0.75 words per token in English. Tokens matter to AI companions because every limit and cost is measured in them — the context window (how much the model remembers), pricing (most APIs bill per token), and response length. When a companion 'runs out of memory,' it's running out of token budget in the window.

What other terms relate to token (ai)?

Related terms in the AI companion space include: Context window, LLM (large language model), Long-term memory (in AI companions). Each has its own glossary entry on /glossary.