Usage Report changedapi/beta/organization/usage_report
Nearest release: v2.1.322, published 2 hours before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+7added
Lines−1removed
From line
44
where the diff opens
First seen
10 Sep 2026
this site's first read of the page
Recorded edits19to this page, all time
The whole hunk
from line 44, old and new numbered
/
from line 44
4444
4545- `context_window: optional array of BetaAnalyticsContextWindow`
4646
47 Restrict usage returned to the specified context window(s).
47 Restrict usage returned to the specified context window(s). See the `context_window` response field for how a request is classified.
4848
4949 - `"0-200k"`
5050
from line 255
255255
256256 - `"webhooks-2026-10-07"`
257257
258 - `"admin-user-sign-out-2026-10-08"`
259
258260 - `"message-threads-2026-08-12"`
259261
260262 - `"analytics-usage-cost-2026-08-24"`
from line 313
311313
312314 Context window used. `null` if not grouping by context window.
313315
316 A request is classified by comparing its prompt size with its model's own long-context boundary, the same boundary that billing uses. The two values are labels for the two classes, not literal ranges: where a model's boundary is below 200k, a request whose prompt exceeds that boundary is reported under `200k-1M`, and a `0-200k` filter does not return it. Where the report cannot determine a model's boundary, it uses 200k instead. Hourly usage stored before this classification took effect keeps its earlier value until it expires. Billed amounts are not affected.
317
314318 - `"0-200k"`
315319
316320 - `"200k-1M"`
from line 924
920924 - `context_window: BetaAnalyticsContextWindow or null`
921925
922926 Context window used. `null` if not grouping by context window.
927
928 A request is classified by comparing its prompt size with its model's own long-context boundary, the same boundary that billing uses. The two values are labels for the two classes, not literal ranges: where a model's boundary is below 200k, a request whose prompt exceeds that boundary is reported under `200k-1M`, and a `0-200k` filter does not return it. Where the report cannot determine a model's boundary, it uses 200k instead. Hourly usage stored before this classification took effect keeps its earlier value until it expires. Billed amounts are not affected.
923929
924930 - `"0-200k"`
925931
No line in this hunk matches that.