Torch not compiled with CUDA enabled

[aimartyr]

I cannot recognize my GPU and my model can only run on my CPU. However, I do have a GPU and I want to utilize it. It seems that if I modify "start-webui.start" to "call python server.py --cai-chat" to utilize my GPU, the following error will occur：
Traceback (most recent call last):
File "F:\AIwebUI\one-click-installers-oobabooga-windows\text-generation-webui\server.py", line 191, in
shared.model, shared.tokenizer = load_model(shared.model_name)
File "F:\AIwebUI\one-click-installers-oobabooga-windows\text-generation-webui\modules\models.py", line 49, in load_model
model = AutoModelForCausalLM.from_pretrained(Path(f"models/{shared.model_name}"), low_cpu_mem_usage=True, torch_dtype=torch.bfloat16 if shared.args.bf16 else torch.float16).cuda()
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 749, in cuda
return self._apply(lambda t: t.cuda(device))
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 641, in _apply
module._apply(fn)
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 641, in _apply
module._apply(fn)
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 664, in _apply
param_applied = fn(param)
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\nn\modules\module.py", line 749, in
return self.apply(lambda t: t.cuda(device))
File "F:\AIwebUI\one-click-installers-oobabooga-windows\installer_files\env\lib\site-packages\torch\cuda_init.py", line 221, in _lazy_init
raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

[aimartyr]

The output shows that you have installed the CPU version of PyTorch, which means it does not support CUDA. You need to install a PyTorch version that supports CUDA and use that version in your code. Installation instructions can be found on the PyTorch official website.If I resolve it, I will close this issue.