Skip to main content

​
Usage

In the dashboard:
  • Overview & Billing show your current 5-hour window, weekly (resets Monday 14:00 UTC), and monthly allowance usage as percentages, along with reset countdowns.
  • Usage also lists the token rates (Input, Output, and Cached Read per 1M tokens) at which each included model counts toward your allowance.
  • Usage & Request Logs record every call with the model, input/output/cached token counts, latency, and cost.
  • Separate tallies track included plan allowance usage vs on-demand credit consumption.
Only success and client_error (4xx) requests count toward your plan allowances. Upstream provider errors and capacity-unavailable responses never count against your allowance.