Get cost by model
Aggregates spend by model over a time range, priced in the cluster currency from the model price book. Org-scoped: org admins see their org; cluster admins may pass org_id. Includes billable work with no usage row of its own (store=false responses, agent tool calls), so totals cover all LLM spend. Usage stays provisional for 24 hours after it occurs: final says whether the whole window is settled, and unpriced_entries counts entries with no cost yet (left out of the totals). Platform-wide, cross-service cost reporting, including compute and sandbox, lives in the budgeting API (/api/v1/budgeting/usage/costs), the cost of record.
Authorizations
Query Parameters
Response
OK
Usage entries in the window with no cost yet, which add 0 to the totals: a missing price or a rating failure, or a response still in progress (priced when it completes, and then counted in its completion window). A response cancelled or failed before it used any tokens costs nothing and is not counted
-9007199254740991 <= x <= 9007199254740991