The atomic unit an LLM reads and generates — roughly ¾ of a word in English.
1 min readupdated 2026-06-22
/ quick answer
A token is a chunk of text (character, subword or word) that an LLM processes. Pricing, context windows and rate limits are all measured in tokens, not characters. The atomic unit an LLM reads and generates — roughly ¾ of a word in English.
The atomic unit an LLM reads and generates — roughly ¾ of a word in English. A token is a chunk of text (character, subword or word) that an LLM processes. Pricing, context windows and rate limits are all measured in tokens, not characters. In practice: 'ChatGPT is great' ≈ 4 tokens. A 4,000-word article is ~5,300 tokens. GPT-4o's 128k context window fits roughly a 300-page book. This dictionary node is part of the Onexial knowledge graph and links to related concepts, workflows and tools below.
Definition
A token is a chunk of text (character, subword or word) that an LLM processes. Pricing, context windows and rate limits are all measured in tokens, not characters.
Example
'ChatGPT is great' ≈ 4 tokens. A 4,000-word article is ~5,300 tokens. GPT-4o's 128k context window fits roughly a 300-page book.
/ frequently asked
What is Token?
A token is a chunk of text (character, subword or word) that an LLM processes. Pricing, context windows and rate limits are all measured in tokens, not characters.
What is an example of Token?
'ChatGPT is great' ≈ 4 tokens. A 4,000-word article is ~5,300 tokens. GPT-4o's 128k context window fits roughly a 300-page book.
Why does Token matter for AI and automation?
The atomic unit an LLM reads and generates — roughly ¾ of a word in English. It connects to the workflows, prompts and tool stacks linked on this page, so you can move from definition to execution without leaving Onexial.