Interesting. I did not know this. Probably because this seems to be confined to the data center boards. MIG for NVIDIA seems confined to the data center chip

Interesting. I did not know this. Probably because this seems to be confined to the data center boards. MIG for NVIDIA seems confined to the data center chips: https://developer.nvidia.com/blog/dividing-nvidia-a30-gpus-and-conquering-multiple-workloads/ https://developer.nvidia.com/blog/dividing-nvidia-a30-gpus-a... For AMD, from your reference, this is Firepro only? Any other VDI references seemed to be data center only as well.