The “Usage” sub-tab under Settings → “Billing & Usage” shows your “Token Usage” — “LLM token consumption across chat, agents, document processing, and more.” On deployments where billing is off, the same view appears as a standalone “Token Usage” tab.
Token Usage overview

Credit meters

When billing is on, meters at the top show your balances:
  • “Your credits” — your allotment this period, with ” used”, when it “Resets”, and a “Next reset” badge.
  • “Your extra credits” — credits an admin assigned to you. These are “Used after your monthly credits run out.”
  • “Credit pool” — an admin-only meter showing the workspace pool.

Key metrics

Four KPI tiles summarize the current period:
  • “Input tokens” — prompt and context sent to the model.
  • “Output tokens” — content generated.
  • “Credits consumed”
  • “Requests”
The “Token consumption” chart plots usage over time. Use the range toggle to switch between 7, 30, and 90 days (30 days is the default), and the metric switch to view Credits or Tokens. Below the chart, a “History” table lists each request — Time / Type / Service / Input / Output / Total / Credits. When a request costs nothing, its credits show as “free”. The table is paginated.
Usage trends chart and history table

Export a CSV

1

Open the export dialog

Click “Export” to open the “Export usage” dialog.
2

Set a date range

Choose “From” and “To” dates.
3

Pick usage types

Use the type checkbox grid to include: chat, agent, super_agent, scheduled_job, workflow_execution, workflow_generation, title_generation, summarization, document_extraction, and image_description.
4

Export

Confirm to download token-usage.csv.

Usage & Quotas

A separate “Quotas” sub-tab covers feature and resource limits — “View your usage and quotas for Generate features and resources.” Use it to see how close you are to your deployment’s configured limits.
For a log of who did what — rather than what it cost — see the separate Activity area, which has its own CSV export.