NVidia CUDA unter WSL für Windows 10/11/12 !
Certainly! To enable NVIDIA CUDA on Windows Subsystem for Linux (WSL), follow these steps:
Install Windows 11 or Windows 10, version 21H2:
Make sure you have either Windows 11 or Windows 10, version 21H2 installed on your system.
Install the GPU driver:
Download and install the NVIDIA CUDA-enabled driver for WSL. This driver allows you to use existing CUDA machine learning workflows within WSL.
For guidance on which driver to install, refer to the Getting Started with CUDA on WSL 2 documentation.
Enable WSL:
After installing the driver, ensure that WSL is enabled on your system.
Install a glibc-based distribution (such as Ubuntu or Debian) within WSL.
Update the kernel:
Make sure you have the latest kernel installed (version 5.10.43.3 or higher).
You can check the kernel version by running the following command in PowerShell:
wsl cat /proc/version
Start using NVIDIA CUDA:
Follow the instructions in the NVIDIA CUDA on WSL User Guide.
You can now use your existing Linux workflows through NVIDIA Docker or by installing PyTorch or TensorFlow inside WSL.
Leave a Reply