YOLOv8 on Jetson Nano

Hi
I have trained a computer vision model on YOLOv8, and I have python 3.6.9 (default) and 3.8.0 in a virtual environment. Since I have Jetpack 4.6.1, can I run YOLOv8 with GPU enabled on Jetson Nano (4GB)? Can Pytorch be enabled on python 3.8.0?

Hi,

You can find the PyTorch package for JetPack 4 below:

Thanks.

I have pytorch 1.10.0 on my jetson nano with python 3.6.9. Can I enable my GPU with python 3.8.0? As the YOLOv8 variant required minimum of python 3.8.0.

Hi,

We only provide the package with the default Python version.
But you can try to build it from the source.

Thanks.

You can use Ultralytics on old Jetson with Docker

Will this method be able run 2 yolov8 models with GPU (Pytorch/CUDA) on videos smoothly at once?

Maybe 1 or 2 FPS if you’re using the nano model.