Degraded inference performance for GLM 5.3

Degraded performance Resolved · Down 29m

Affected: inference

Lifecycle

1. Reported 23 Sep, 15:59 UTC operator
2. Confirmed not reached
3. Investigating 23 Sep, 15:59 UTC operator
4. Resolved 23 Sep, 16:29 UTC operator
5. Closed not reached

Timeline

1. Resolved 23 Sep 2026, 16:29 UTC operator
   GLM 5.3 inference performance has returned to normal at inference.pearlresearch.ai. We are continuing to monitor for any recurrence.
2. Reported 23 Sep 2026, 15:59 UTC operator
   Degraded inference performance has been reported for GLM 5.3 at inference.pearlresearch.ai.
3. Investigating 23 Sep 2026, 15:59 UTC operator

   GLM 5.3 is experiencing degraded performance on our inference endpoint at inference.pearlresearch.ai. We are investigating and will share updates as more information becomes available.

   We apologize for the disruption and thank you for your patience.