[Question]CUDA error: no kernel image is available for execution on the device

windows
1

Hi, I am trying to install and use PyTorch on my machine, but something went wrong:

AcceleratorError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

My system specifications are given below:

OS: Windows 10
Cuda: Cuda compilation tools, release 12.6, V12.6.85
Device name: NVIDIA GeForce RTX 5090 D
torch version: 2.8.0+cu126
python version: 3.1.1

Can someone help me solve this problem? Thank you so much!

2

Uninstall the current PyTorch binary and install any stable or nightly PyTorch binary built with CUDA 12.8+ which is supporting your Blackwell GPU.