Token Counter

Estimate how many tokens a piece of text will use in a language model, from its characters and its words. It runs in your browser.

How to use it

1. Paste your text.
2. Read the two estimates, the character count and the word count.
3. Use the higher estimate when you are close to a limit.

How the estimate works

For English text, a token averages about four characters or three quarters of a word. The tool divides characters by 4 and words by 0.75. “The quick brown fox jumps over the lazy dog.” (44 characters, 9 words) gives about 11 tokens by characters and 12 by words.

Good to know

This is an estimate. It does not run a model’s real tokenizer, and different models tokenise differently. Code, numbers and unusual words use more tokens per character, and many non-English languages need more tokens for the same meaning. For exact counts, use the tokenizer provided by your model’s maker. To turn a count into a cost, use the LLM Cost Calculator.

See the guide and the token glossary entry.

FAQs