The issue has been resolved.
Thank you for the patience.

Some models are intermittently slow or timing out due to a network issue at one of our infrastructure providers. The issue comes in short bursts, and affected requests recover automatically once each burst passes. We are working with the provider on a fix and will post updates here.

We’ve migrated some of the consumers to a stable location. Queues are back to normal, we will keep monitoring this incident for a while

We've identified the issue with our third-party provider and now preparing the remediation.

The team is still investigating