LLM Token Counter and Estimator
Paste text to see characters, words and an approximate token count.
- Approximate tokens
- 0
- Likely range
- 0 to 0
- Characters
- 0
- Characters without spaces
- 0
- Words
- 0
- Lines
- 0
How the estimate works
Language models read text as tokens: pieces of words, whole short words, punctuation or spaces. Each model family uses its own tokenizer, so the same text can produce different counts. This tool does not run a real tokenizer. It averages two rules of thumb for English prose: about 4 characters per token and about 0.75 words per token. The range shows both.
Code, non-English text and unusual formatting often use more tokens than this estimate. For billing-critical numbers, use your provider's own token counting tool.
How to use it
- Paste your text, or choose Load example.
- Read the approximate token count and the likely range.
- Use Copy results to save the numbers, or Reset to start over.
Example
A 400-character paragraph of about 70 words gives roughly 100 tokens by the character rule and about 93 by the word rule, so the estimate is around 97.
Frequently asked questions
Is this the exact token count?
No. It is an estimate based on character and word counts. Exact counts depend on the specific model's tokenizer.
Why do code and other languages count differently?
Tokenizers are usually tuned on common English text, so code, numbers and many non-English languages are often split into more tokens per character.
Is my text sent anywhere?
No. Counting happens in your browser.