l want to ensure that the deepness has the support for yolov11?

[Marchis108]

l input my yolov11 onnx model,and while its runing , my cpu is full ,and the GPU is empty.

[bartoszptak]

Hi @Marchis108,
Yes - YOLOv11 is supported.
To utilize GPU inference, you must configure your computer correctly to meet all ONNXRuntime-GPU requirements.

For example, on Ubuntu, you need to set variables:

export CUDA_HOME=/usr/local/cuda
export LD_LIBRARY_PATH=/usr/local/cuda-12.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
export PATH=/usr/local/cuda-12.1/bin${PATH:+:${PATH}}

[Marchis108]

Hi @Marchis108, Yes - YOLOv11 is supported. To utilize GPU inference, you must configure your computer correctly to meet all ONNXRuntime-GPU requirements.

For example, on Ubuntu, you need to set variables:

export CUDA_HOME=/usr/local/cuda
export LD_LIBRARY_PATH=/usr/local/cuda-12.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
export PATH=/usr/local/cuda-12.1/bin${PATH:+:${PATH}}

could l use gpu to train the model on Windows？

[bartoszptak]

Generally, ONNXRuntime and Deepness are not suitable for model training. Check libraries such as PyTorch or Ultralytics.

I've never tried using ONNXRuntime-GPU on Windows or macOS either.