The Minisforum MS-A2 has been my go to platform for running an optimized and fully functional VMware Cloud Foundation (VCF) 9.0 lab environment. From vSAN ESA to NSX VPC to vSphere Kubernetes Service to VCF Automation to Data Services Manager, it runs it all!
Over the past couple of weeks, I had also been experimenting with running VMware Private AI Services (PAIS) using the MS-A2. For those with an eye for detail, may have noticed that the NVIDIA GPU requirement was actually being satisfied by an ASUS NUC 14 Performance, which includes an NVIDIA GeForce RTX 4070 (8GB VRAM). The ASUS NUC was purely for validation and prototyping purposes to ensure that I could actually run PAIS before investing in a more capable and certainly more expensive NVIDIA GPU ๐ค
One of the lesser known capabilities of the MS-A2 is the ability to add a half-height PCIe device and from the very beginning when I had first heard about the MS-A2, my plan and hope was to be able to add an NVIDIA GPU to the system!
With the compact form factor of the MS-A2, you are definitely limited by the types of PCIe devices that can be used and thats when I came across custom GPU coolers from n3rdware. Both the Minisforum MS-01 (Intel variant) and the MS-A2 can support NVIDIA RTX A2000, RTX 2000 Ada and RTX 4000 Ada.
I opted for the most capable GPU for running AI workloads and went with an RTX 4000 SFF Ada and with the largest amount of VRAM which was 20GB, definitely not a cheap purchase! ๐
- RTX 4000 SFF Ada with 20GB VRAM (VCNRTX4000ADALP-PB)
- RTX 4000 SFF Ada custom cooler
- Arctic Silver 5 Thermal Paste
Note: If you decide to buy a different GPU, just make sure you purchase a compatible cooler and that the GPU is matching what is supported since some of the online listings, its not always clear if it is single-slot or full desktop card. This is also why I listed the specific SKU for the RTX 4000 SFF Ada w/20GB VRAM. If you are interested in the other GPU cooler options, I have links to respective options for RTX A2000 & RTX 2000
The installation was super simple, you can either view the instructions from n3rdware or if you prefer watching a video, which I found quite informative in properly taking apart the GPU.
Here is a picture comparing the stock GPU shroud and the n3rdware custom cooler

Here is a picture of the raw GPU card, it is pretty insane how much of the GPU mass is actually made up of the heatsink, fan and shroud

Here is a picture of the final product, super slick looking and fits perfectly in the MS-A2!

๐
The moment of truth was to power on the MS-A2 and ensuring nothing fried and that the GPU fan was running and ESXi was able to properly detect the NVIDIA GPU! ๐



how are use this with vgpu on esxi ?
Iโm not, itโs used in passthru