with that said am i the only one that it seem the response is a lot quicker and the gpt 4.0 model is doing better with responses? not sure what has change recently