Hello @Scheffel, Bruce
I had a quick discussion with the product team on your ask. The cards mentioned below are offered to cloud providers and many OEMs do not offer them in their systems because of the large size and usage. The GB200 is also currently offered to cloud providers as it’s a CPU/GPU chip.
Latest cards supported for GPU-P/vGPU are A2, A16, A40, L4, L40, L40S(supported by NVIDIA in v17 and the upcoming v18 driver versions).n Other GPUs are supported on DDA mode.
It would be more helpful to understand the workloads that are intended for the GPUs in question or if you are just looking for a check-list for support.
As mentioned above, there is a range of GPUs offered today - and limitations on other models that would not be directly available, but we still want to meet customer needs.
Hope this helps!
If I have answered your question, please accept this as answer as a token of appreciation and don't forget to thumbs up for "Was it helpful"!