On this page
Console view
Console › Usage has an Overview and a Request log, both filterable by time range (last 7 / 30 / 90 days) and API key:
| View | Question it answers |
|---|---|
| By time | Daily volume and cost trend — where the spikes are |
| By key | Which workload spends most, and whether a key is being misused |
| By model | Where cost concentrates, and what could move to a smaller model |
| Request log | Each call's model, usage and cost, searchable by request ID |
Per-call detail
The usage object in each response is the ground truth:
json
{
"usage": {
"prompt_tokens": 1820,
"completion_tokens": 412,
"total_tokens": 2232,
"prompt_tokens_details": { "cached_tokens": 1536 }
}
}cached_tokens is input served from prompt cache and billed at the lower cache-read rate — see billing.
Note
Usage data can lag by about five minutes and is not guaranteed real-time. Reconcile against the next day's figures.
Contact us
Something unclear, or need a human? Pick the channel that fits your question.
Supportsupport@apione.ai
WeCom
Telegram
Communityhttps://t.me/apione
Support tickets
Billing & accountConsole › Feedback