Container Images
To use a GPU from a running container, the host system must be set up with a kernel-mode driver that supports your GPU. To check the upstream kernel version in which support for a given GPU was first enabled, refer to the hardware table. Treat this as guidance only; vendor kernels may include backported support for newer hardware in older kernel versions. Always refer to your operating system release notes to confirm which Intel GPUs are supported in your operating system release.
To verify that your operating system can detect your GPU, follow the host system setup instructions in the installation procedures. To pull prebuilt container images from Docker Hub or build them from Dockerfiles on GitHub, use the links in the following table.
The provided container images are based on Ubuntu and are not intended for production use. To receive expanded security maintenance from Canonical on the Ubuntu base layer, you can follow the Ubuntu guide to enable Ubuntu Pro in a Dockerfile, which requires rebuilding the image.
Image |
Description |
Variants |
DockerHub |
Dockerfiles |
|---|---|---|---|---|
compute-runtime |
The Intel® Graphics Compute Runtime for oneAPI Level Zero and OpenCL™ Driver is an open source project providing compute API support (Level Zero and OpenCL) for Intel graphics hardware architectures (HD Graphics and Xe). |
runtime, devel |
||
intel-omix |
Intel® Open Middleware Xe (Intel® OMIX) is a set of highly optimized deep learning frameworks and tools for Intel GPUs to accelerate AI workloads. |
runtime, devel |
||
pytorch |
Intel GPU support for PyTorch for Intel® Client GPUs and Intel® Data Center GPU Max Series, which brings Intel GPUs and the SYCL software stack into the official PyTorch stack with consistent user experience to embrace more AI application scenarios. |
cpu, xpu |