Skip to main content

Activity

Your usage across models over the last 7 days or 30 days.

The Activity page: totals, token breakdown, daily spend, top models and keys

What you see​

  • Spend, Requests (split into ok and failed), Tokens and Avg / request for the period.
  • Input tokens, Output tokens, Cache read tokens and Cache write tokens. Cached prompt tokens are the part of a repeated long prompt the model served from its cache; they are billed at the model's cached rate.
  • Daily spend: one point per day. Days with no requests show as zero, so gaps are real gaps.
  • Top models and Top Virtual Keys: where the spend went, by model and by key.

Typical use: find what a spike cost​

  1. Find the day on the Daily spend line.
  2. Check Top models and Top Virtual Keys for the same period to see which model and which application drove it.
  3. Open Logs for that day to see the individual requests.