I am trying to setup torch module in my windows gpu server but its failing with below error.

gpu server details:
nvidia-smi

| NVIDIA-SMI 551.86 Driver Version: 551.86 CUDA Version: 12.4

error,
raise err
OSError: [WinError 126] The specified module could not be found. Error loading “C:UsersuserAppDataLocalProgramsPythonPython311Libsite-packagestorchlibshm.dll” or one of its dependencies.

reference:tried below link options also but still its not working.
PyTorch Error loading “libsite-packagestorchlibshm.dll” or one of its dependencies

i have installed torch,cudatoolkit, ..etc using python and conda but its keep throwing below error. cuda available is always false. even with torch torch==2.2.1 its not working and other modules are not working with this versions. my gpu versions are as below .
cmd: pip3 install torch torchvision torchaudio –index-url https://download.pytorch.org/whl/cu118