I’m comparing the Azure OpenAI service with the OpenAI API platform and have some concerns regarding cost management.

In the OpenAI API platform, there is a prepaid model, and spending is limited to that amount. In contrast, Azure operates on a pay-as-you-go model. Even if you set up budgets and alerts in Azure, there’s a risk of quickly incurring high costs (e.g., by using very long prompts) before you have a chance to react.

Does it make sense to use Azure OpenAI given this potential risk? How do you effectively manage and limit costs in Azure to prevent unexpected high charges?

I have looked into setting budgets and alerts in Azure to monitor spending. While these tools are helpful, they don’t fully mitigate the risk of someone running up large costs in a short period, such as by submitting very long prompts to the Azure OpenAI service.