How many tokens
is this text?
Paste a prompt or document and get a fast estimate of its token count and input cost.
Token estimator.
A quick estimate in your browser. Nothing you paste is sent anywhere.
- Likely range
- 0 – 0
- Characters
- 0
- Words
- 0
- Input cost once · Claude Sonnet 5.5
- USD 0.00
- Input cost × 1,000 requests
- USD 0.00
Uses the published rule of thumb of about four characters per token, or 60–80 English words per 100 tokens. For billing-grade counts, use the provider’s token-counting API with your exact model.
Every model counts differently.
Each model family splits text with its own tokenizer. Anthropic notes that its newer tokenizer produces roughly 30% more tokens for the same text than earlier Claude models. Languages written in non-Latin scripts, code and unusual formatting often use more tokens per character than plain English.
Read: why the same text costs differently →Ask the provider.
For a billing-grade count, use the provider’s own tool. Anthropic’s token-counting endpoint is free to use and accepts the same input as a message. Gemini offers a count_tokens method and returns usage on every response. Every major API reports input and output tokens in its response.