We’re adding support for GPU node pools in AKS (Azure Kubernetes Service). GPU nodes are great for compute-intensive workloads such as graphics and visualization workloads, or machine-learning processes. Azure uses the NVIDIA device plugin to make the GPU capacity of a node available to Kubernetes workloads.
You can find all the virtual machine sizes available in Azure here.
To know more about how to schedule GPUs for your workloads, head to the official Kubernetes documentation on the topic.