[CSTrack] What is the difference between the yolov5l.pt you provide and the yolov5l.pt officially provided by yolov5

Thank you for your contribution. I would like to know what is the difference between the pre-trained model YOLOV5.pt provided by you and the official model provided by yolov5, and whether the model provided by you can adapt to the image input [1088,608]. If so, I would like to know how this is achieved. Thank you very much