Still using it but the regions issue is annoying. We currently deploy to an entirely odd region way outside our geographical location because it had the most a

Still using it but the regions issue is annoying. We currently deploy to an entirely odd region way outside our geographical location because it had the most available models when we set up, even then it still lagged behind and only got gpt-4o recently. I find the whole concept of a "deployment" odd, clearly it's not deploying anything other than a config enabling access for your account. Great, now I have to keep track of every model release and have a DevOps process to keep on track! Also yes we noticed slow downs, but as we've stopped using openai directly I don't know if this just applies to azure.