Usage
Inspect Tokener.ai usage by day, model, and API key.
Usage in the console is imported from the model gateway and rolled up for dashboards. You can filter spend by day, model, and API key.
What is tracked
- Request counts
- Prompt and completion tokens
- Cost in credits
Freshness
Metering settles within minutes of each call under normal conditions. If a chart looks stale, refresh after a short wait — imports and rollups run asynchronously.
Raw vs rollups
Dashboards read aggregated rollups, not raw gateway event tables. That keeps the console fast as volume grows.
Request logs
Logs shows per-request records from the last 7 days. Filter by a 24-hour or 7-day range, request ID, status, API key, or model. Records normally appear within about two minutes and include token counts, cache status, and cost.