Token

What is Token?

A token is the fundamental unit of text processing for Large Language Models (LLMs). Rather than processing whole words or individual characters, models break text into 'tokens' which can represent words, sub-words, or punctuation. For English, 1,000 tokens corresponds to roughly 750 words.

Where did the term "Token" come from?

Derived from lexical analysis in computer science and linguistics.

How is "Token" used today?

The universal currency for measuring LLM context windows, computational cost, and API pricing.

Related Terms