Win10 + RTX 4080

I have the follow result running python -m torch.utils.collect_env:

PyTorch version: 1.13.1+cpu
Is debug build: False
CUDA used to build PyTorch: Could not collect
ROCM used to build PyTorch: N/A

OS: Microsoft Windows 10 Pro
GCC version: Could not collect
Clang version: Could not collect
CMake version: Could not collect
Libc version: N/A

Python version: 3.9.13 (tags/v3.9.13:6de2ca5, May 17 2022, 16:36:42) [MSC v.1929 64 bit (AMD64)] (64-bit runtime)
Python platform: Windows-10-10.0.19045-SP0
Is CUDA available: False
CUDA runtime version: 12.1.66
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4080
Nvidia driver version: 531.14
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

Versions of relevant libraries:
[pip3] lion-pytorch==0.0.7
[pip3] numpy==1.24.1
[pip3] PaLM-rlhf-pytorch==0.0.65
[pip3] torch==1.13.1
[pip3] torchaudio==0.13.1+cu117
[pip3] torchvision==0.14.1+cu117
[conda] Could not collect

When I run (torch.cuda.is_available()) it returns False. I see that I have installed pytorch 1.13.1+cpu, but after numerous reinstalls it is still 1.13.1+cpu. How do I change it to not be +cpu?

Welcome to the forums!

I have used that already, but it still returns the same result for unknown reasons :confused:

If you initially installed the cpu version, you may need to completely uninstall torch, torchvision and torchaudio first.

I am about to order a laptop that contains nvidia rtx 4080. is it compatible with pytorch? I would appreaciate if someone would share a link where the supported rtx models by pytorch are listed.

Yes, all current PyTorch binaries support the 4080.

1 Like

is there a specific page for tracking the latest developments in terms of hardware?