Load balancing Azure OpenAI

Normal scenario:

• token and model quota is available
• api return 200 Throttled scenario:
• quota limits is applied
• api managment retry to a different resource

References

https://learn.microsoft.com/en-us/azure/developer/python/get-started-app-chat-scaling-with-azure-api-management?tabs=github-codespaces%2Cinitial-deployment