The response time of open ai models is increased and taking more time for MODEL = “gpt-3.5-turbo” and MODEL = “gpt-4o-mini” and gpt-4o-mini starts working in between gives results and then stop and lot of time. gpt-3.5-turbo was taking earlier 3-4 seconds now taking 40-60 seconds to give the response.
1 Like
Where are you making the calls from and with what service?
1 Like
through prompting earlies it takes 2-5 seconds with same prompt but now it is taking 40-50 second to send the response.
1 Like
I understand this part.
I am wondering how you are making this call. Are you doing it on your computer? Through a service? Where is the call coming from (country-wise)?
1 Like
On my machine and in country india
1 Like
Thanks. Which platform are you using to make these calls? A jupyter notebook? Do you have an always-online program that makes them?
If you go to the OpenAI playground and use the same prompt, what happens?
1 Like