In the Dell Technologies HPC & AI Innovation Lab, we deployed and validated the solution described in this design guide. We deployed NVIDIA AI Enterprise on supported PowerEdge Servers. We conducted experiments to validate the training and inference performance. The studies were conducted on VMs and Tanzu Kubernetes clusters to characterize the performance of partitioned GPUs. This section presents the results of our performance studies and our validation of an AI radiologist use case using cnvrg.io.