Does the Rate limit (Tokens per minute) of an Azure OpenAI instance corresponds to the input tokens, the output tokens or the sum of both?

This question has been addressed elsewhere.

This question is not within the scope of Software Development.

This question cannot be answered in its current form, because critical information is missing.

This post contains multiple questions or has many possible indistinguishable correct answers or requires extraordinary long answers.

This question cannot be answered in a way that is helpful to anyone. It's not possible to learn something from possible answers, except for the solution for the specific problem of the asker.

Code posted for review must be complete and working.