Token Counter
Estimate tokens and API cost before you send a prompt.
Estimated cost per model tier
Assumes this text as input and a reply roughly one quarter its length. Provider prices change often — verify before budgeting.
| Model tier | Per run | 1 runs |
|---|---|---|
| Frontier tier (e.g. GPT-class, Claude Opus-class) | $0.00 | $0.00 |
| Balanced tier (e.g. Sonnet-class, GPT mid-tier) | $0.00 | $0.00 |
| Fast tier (e.g. Haiku-class, mini models) | $0.00 | $0.00 |
About this tool
Paste a prompt or document to estimate how many tokens it uses and what it would cost across common models. Useful for budgeting API work and for checking whether a long document will fit in a model's context window.
Tokens are the units models actually read — roughly three-quarters of a word in English, but far denser for code, punctuation and non-English scripts. This tool estimates rather than tokenizes: real tokenizers are model-specific and would add megabytes to the page. For billing decisions, check your provider's own usage dashboard.
Do this at scale with AI
This app handles the one-off job. These tools handle it as part of a real workflow.