Calculator › Token counter
Token counter
Paste any text to count its tokens and see what it costs to send to (or get back from) popular models. Counting happens in your browser; nothing is uploaded.
| Model | Tokens | Cost as input | Cost as output |
|---|
Token counts use OpenAI's o200k_base encoding (GPT-4o, GPT-4.1, GPT-5, o-series). Counts for other providers are estimates; Claude 4.7+ counts are multiplied by 1.3 to reflect Anthropic's newer tokenizer. Standard, uncached prices.
What is a token?
Language models read and write text in chunks called tokens. A token is often a whole common word, part of a longer word, a number group or a punctuation mark. As a rule of thumb, one token is about 4 characters or ¾ of a word of English, so 1,000 tokens is roughly 750 words. Chinese, Japanese and Korean text usually uses about one token per character or more, and source code often uses more tokens than prose because of symbols and indentation.
API providers bill per token, separately for the tokens you send (input) and the tokens the model generates (output). That is why the same prompt can cost very different amounts on different models, and why output length matters so much.
Rules of thumb
- 1 page of a typical report ≈ 500–700 tokens
- A 30-minute meeting transcript ≈ 5,000–7,000 tokens
- A 300-page book ≈ 120,000–150,000 tokens
- Tool definitions and system prompts count as input on every request unless cached
Turn token counts into a monthly budget with the calculator →