Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

I am playing around with YOLOv5 and exporting to different formats (.pt & .torchscript). But I found that after I exported the pre-trained models using this command:
!python "/content/drive/MyDrive/yolov5/export.py" --weights "/content/drive/MyDrive/yolov5/yolov5s.pt" --include torchscript

The .torchscript model's bounding is off, please advise me on why this is happening and how to resolve it. Thank you.

.pt model:

.torchscript model:

Additional

No response