Tool-calling models are priced 3103× apart for the same declared capability.

Across 274 models that declare tool support, blended prices run from $0.0218 to $67.50 per million tokens. Both figures come from the same held-out agentic tool use suite, 274 scored items, measured the same way — so the comparison is like for like.

Price is set by what a provider can charge, not by what a model scores. 3103× is what that gap costs a buyer who picks on reputation.

Someone is paying the difference. If that is you, the question worth asking is which of these two numbers your workload actually needs — and this is the measurement that answers it.

2 measurement cycles ran in the last 24 hours, 2 newly listed models were measured, no recipe reached a frontier.

The current numbers live on the measured answers; the method is public.