Know what your LLM calls actually cost.
Free tools for comparing API pricing and measuring token usage across 45 models from 5 providers. Prices come straight from official provider documentation and are re-verified on a schedule — last checked 2026-08-18.
No account. No upload. Everything runs in your browser.
Why another pricing calculator?
Because most of them are wrong in the same three ways: they go stale within weeks of a price change, they ignore prompt caching and batch discounts, and they assume a token means the same thing on every model. It does not — Claude 4.7 and later produce roughly 30% more tokens for identical text than earlier Claude models, which is enough to invert a ranking. TokenGauge models all of it, cites the official source for every number, and leaves a field blank rather than publishing a figure it cannot verify.