Token EstimatorHow to use it
- Paste your text. You get a range of tokens, from the two rules of thumb.
- Type the context window of your model to see if it fits, keeping some space for the answer.
- Type the prices from the maker's pricing page to see the cost of one request.
What the range means
The low and high numbers come from the Gemini documentation's rules: about 4 characters per token, and 100 tokens for 60 to 80 English words. The real count of your model will usually lie near or inside the range for ordinary English, and above it for many other languages.
Is my text stored?
No. It stays in your browser.
Why no price list?
Prices change often, and a wrong price would give a wrong budget. You type the current price.
This is an independent guide. It is not connected with Google, Anthropic, OpenAI or any company named here. Facts were read on 14 November 2026 from the pages named on each page. The numbers are estimates from rules of thumb, not the counts of any maker's tokenizer.