Degraded performance on uGPT replies

We have not observed any errors, and all of our services are functioning normally.

We are observing that the overall performance is a lot better already, most of the requests are successful and with the expected latency.

We'll still monitor the situation while the cluster continues escalating.

The infrastructure is currently escalating and we expect this to take a couple of hours more to complete.

We'll be monitoring how it affects the performance, for now we continue to see degraded performance for some bots.

We are seeing occasional 503 errors from our provider, as a first measure we increased our service tier and we are trying to connect with their support team.

We are currently investigating an issue with uGPT replies which seem to be failing in some scenarios.