Some systems degraded

We have identified the root cause — an upstream model provider is experiencing elevated queue times. A mitigation is in place and we are continuing to monitor response times.

We have identified the issue as being related to our primary inference provider. We are routing a portion of traffic to the secondary provider.

We are investigating reports of slow AI responses. All other services are unaffected.

Scheduled maintenance window has started. The web dashboard may be intermittently unavailable for up to 2 hours while we upgrade our database cluster.