Once we have both a cluster deployed with GPUs exposed and OpenShift AI without GPU acceleration, we can work to combine the two and enable GPU acceleration. The documentation for doing this can be found here:

https://access.redhat.com/documentation/en-us/red_hat_openshift_ai_self-managed/2-latest/html/installing_and_uninstalling_openshift_ai_self-managed/enabling-gpu-support_install