pseudo minnow
Hi team! I'm using gpt-3.5-turbo-0301, but I've noticed a significant increase in response times. I'm parallelizing 20 API calls, but even with this approach, 1,300 calls are taking approximately 1.5 hours with 10 tokens of output. Is anyone else experiencing this prolonged response time as well? I consistently receive a message saying "The server is overloaded or not ready yet."
Do you have any recommendations?