diff --git a/changelog/fragments/cooldown-per-model-stepped-backoff.md b/changelog/fragments/cooldown-per-model-stepped-backoff.md new file mode 100644 index 00000000000..228bf0c6737 --- /dev/null +++ b/changelog/fragments/cooldown-per-model-stepped-backoff.md @@ -0,0 +1 @@ +- Agents/cooldowns: scope rate-limit cooldowns per model so one 429 no longer blocks every model on the same auth profile, replace the exponential 1 min → 1 h escalation with a stepped 30 s / 1 min / 5 min ladder, and surface a user-facing countdown message when all models are rate-limited. (#49834) Thanks @kiranvk-2011.