Question:
I’m experiencing difficulty using my NVIDIA Quadro M2000M as a GPU for PyTorch on Windows 10 64-bit. I have installed CUDA 12.5, and when I check the CUDA version using the command nvcc –version, I get the following output:

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Thu_Jun__6_03:03:05_Pacific_Daylight_Time_2024
Cuda compilation tools, release 12.5, V12.5.82
Build cuda_12.5.r12.5/compiler.34385749_0

Additionally, my PyTorch version is 2.3.1+cpu. However, when I tried to install PyTorch with GPU support using the command provided by PyTorch (conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia), I encountered an error. Here is the error message I received:

Downloading and Extracting Packages
pytorch-2.3.1 | 1.39 GB | ###########################################################################9 | 100% ERROR: Encountered corrupt package tarball at C:UsersasmsAnaconda3pkgspytorch-2.3.1-py3.8_cuda11.8_cudnn8_0.tar.bz2. Conda has left it in place. Please report this to the maintainers of your package. For the defaults channel, please report to https://github.com/continuumio/anaconda-issues
pytorch-2.3.1 | 1.39 GB | ############################################################################ | 100%

CondaSystemExit: Exiting 1

I have two questions:

What version of PyTorch is compatible with CUDA 12.5? I want to ensure I install the correct version to work with my CUDA installation.
When I check for available GPUs using PyTorch, I get none. How can I activate my NVIDIA Quadro M2000M GPU for use with PyTorch?
Any guidance on resolving these issues would be greatly appreciated. Thank you!