gpt40613Overrun problem,Which technical expert can help me solve it

Why does my gpt-4-0613api always prompt me to use beyond the range, but actually I have not exceeded 250,000 tokens per minute at all. Does anyone know this problem a