CUDA not able to loaded

[abdulghani91]

When I try to choose any auto model I get the error below:

Error in loading model: D: \a\_work\1\s\onnxruntime\python\onnxruntime pybind state.cc: 537 onnxruntime::python::CreateExecutionProviderInstance CUDA_PATH is set but CUDA wasn't able to be loaded. Please install the correct version of CUDA and cuDNN as mentioned in the GU requirements page (https://onnxruntime.ai/docs/reference/execution-providers/CUDA-ExecutionProvider.html # requirements), make sure they are in the path and your GPU is supported.

I install the executable file and try with Conda and got the same error my system is Windows 11 with GPU RTX 3080 python 3.10.8

nvcc --version nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2022 NVIDIA Corporation Built on Wed_Jun__8_16:59:34_Pacific_Daylight_Time_2022 Cuda compilation tools, release 11.7, V11.7.99 Build cuda_11.7.r11.7/compiler.31442593_0

torch.cuda.is_available() True

Could anyone help me with this problem?

[vietanhdev]

Please check the version of CUDA and CUDNN:

• CUDA must be 11.6.
• CUDNN: 8.2.4 (Linux), 8.5.0.96 (Windows). More information: https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html.

[abdulghani91]

@vietanhdev Thank you.

[doantientai]

Please check the version of CUDA and CUDNN:

• CUDA must be 11.6.
• CUDNN: 8.2.4 (Linux), 8.5.0.96 (Windows). More information: https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html.

Hi @vietanhdev, thanks for your wonderful work! Just to make it clear, CUDA must be 11.6 or >= 11.6 ? I have CUDA 12.1 and the gpu is not used during labeling.