RTX 5060 Ti can not generate image.

[T-Natthaphol]

Got the error below, How can I fix it?

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.

System Info:

GPU: Zotac RTX 5060 Ti 16GB

CUDA: 12.9

Python: 3.10.6

PyTorch: 2.5.1+cu121

OS: Windows 11

[SuperPat45]

Same with RTX 5090,

This error appear because RTX5000 need cu128

Code need to be upgraded to Pytorch 2.7.0 with CUDA12.8

[MisterChief95]

Got the error below, How can I fix it?

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.

System Info:

GPU: Zotac RTX 5060 Ti 16GB

CUDA: 12.9

Python: 3.10.6

PyTorch: 2.5.1+cu121

OS: Windows 11

Open a terminal or powershell window in the Forge directory, and run these commands.

venv/Scripts/activate
python -m pip uninstall torch torchvision torchaudio
python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128

if it still doesn't work, go ahead and add your startup terminal output and we'll look at it more.

[T-Natthaphol]

Got the error below, How can I fix it? 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. System Info: GPU: Zotac RTX 5060 Ti 16GB CUDA: 12.9 Python: 3.10.6 PyTorch: 2.5.1+cu121 OS: Windows 11

Open a terminal or powershell window in the Forge directory, and run these commands.

venv/Scripts/activate python -m pip uninstall torch torchvision torchaudio python -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128 if it still doesn't work, go ahead and add your startup terminal output and we'll look at it more.

Thank you, [MisterChief95]

[BrickDesignerNL]

Additionally you might want to install cuDNN:

To install cuDNN for CUDA 12, run: venv/Scripts/activate python -m pip install nvidia-cudnn-cu12

Source: https://docs.nvidia.com/deeplearning/cudnn/installation/latest/windows.html#installing-cudnn-with-pip

If you get this error

Could not locate cudnn_graph64_9.dll. Please make sure it is in your library path! after generating an image.