fair siren
On average, the response times for the following models from my experience
gpt3.5turbo - 20 sec
gpt4 - more than 60 sec
I did some research and it seems like most people are getting the 3.5 turbo responses within <10 secs. So why is my response so slow? 😦
I am accessing the API through python and the token length is only around 1 thousand. I suspect it is due to my location as I am based in Singapore. Does anyone know why the location would have such a big impact on the response times? Or if there is anyways I can get around this.
wheat dirge