I can’t use my 4090 laptop for pytorch. I followed the instructions for installing CUDA and even contacted Nvidia customer support but when I run:
import torch
print(torch.cuda.is_available())
I get a false statement, showing that torch can’t find my gpu with cuda. Here is what the installer looked like:
For reference, the drivers on my gpu are 537.47 and the cuda version is 537.13.
I also can’t find the CUDA toolkit as an application in my files.
Any help would be appreciated.
Update:
I ran python -m torch.utils.collect_env and I saw that CUDA is there, but it is not available
OS: Microsoft Windows 11 Home
GCC version: (Rev6, Built by MSYS2 project) 13.1.0
Clang version: Could not collect
CMake version: version 3.27.5
Libc version: N/A
Python version: 3.11.5 | packaged by Anaconda, Inc. | (main, Sep 11 2023, 13:26:23) [MSC v.1916 64 bit (AMD64)] (64-bit runtime)
Python platform: Windows-10-10.0.22621-SP0
Is CUDA available: False
CUDA runtime version: 12.2.140
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: GPU 0: NVIDIA GeForce RTX 4090 Laptop GPU
Nvidia driver version: 537.13
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
CPU:
Architecture=9
CurrentClockSpeed=2200
DeviceID=CPU0
Family=207
L2CacheSize=32768
L2CacheSpeed=
Manufacturer=GenuineIntel
MaxClockSpeed=2200
Name=13th Gen Intel(R) Core(TM) i9-13900HX
ProcessorType=3
Revision=
Versions of relevant libraries:
[pip3] numpy==1.23.5
[pip3] torch==2.0.0
[pip3] torchvision==0.15.1
[conda] blas 1.0 mkl
[conda] mkl 2023.1.0 h6b88ed4_46357
[conda] numpy 1.23.5 pypi_0 pypi
[conda] pytorch 2.0.1 py3.11_cpu_0 pytorch
[conda] pytorch-mutex 1.0 cpu pytorch
[conda] torch 2.0.0 pypi_0 pypi
[conda] torchvision 0.15.1 pypi_0 pypi
rs277
September 24, 2023, 9:56pm
3
I know nothing about pytorch, but looking at the homepage, they seem to offer builds for Cuda 11.7 and 11.8 only, not the 12.2 that you have installed.
11.8 would be the version you want, as that was when support for your GPU was added.
THANK YOU!
Fwiw and anyone having this issue: re-download pytorch for 11.8 Start Locally | PyTorch
Then redownload cuda for 11.8