Elevated latency + error rate on text-davinci-003 and gpt-3.5-turbo

We believe we have identified the issue causing the elevated latency and error rates, and have taken the appropriate mitigations.

We are continuing to investigate underlying latency issues and error rates.

We have identified the cause of the error rate and are deploying a fix

We are observing a 5-10% error rate on text-davinci-003 and gpt-3.5-turbo and approximately doubling of latency. We are continuing to investigate the root cause.

We're seeing elevated error rates and latency on text-davinci-003 and gpt-3.5-turbo. We're investigating the issue