Where do these models report cached prompt tokens?
All five report cache hits under usage.prompt_tokens_details.cached_tokens, and that is the field we bill from.
What we found
Prompt caching only saves you money if you can see it, and providers disagree about where to put the number. Some report it as a separate cache field, some nest it under prompt token details, some report nothing at all and leave you to infer it from the bill.
We checked each upstream individually rather than assuming they matched. All five models served here report cache hits in the same place, under prompt_tokens_details.cached_tokens on the usage object, and every one of them does report it. That is the field our metering reads, so the number in your own response body is the number your invoice was computed from.
This matters more than it sounds, because caching is the largest single lever on what an agent costs.
How this was measured
An unattributed measurement is a marketing claim, so here is the receipt.
- Probe
- probe_cache_fields.py
- Measured
- 2026-09-08
- Applies to
- every model we serve
We found this by running the probe against the models we serve, which is also why it is worth trusting: it is a reading rather than a restatement of somebody's documentation. The full catalog is on /models, and what a request costs is on /pricing.
Upstream behaviour changes without notice, and a measurement is only as good as its date. If you find this is no longer true, tell us on the support page and we will re-run the probe.