GPU ALLOCATION ISSUE TO VM's WITH WINDOWS 11 PRO OS

Ino Sone 0 Reputation points
2026-08-24T23:43:44.4433333+00:00

Hi Microsoft Community,

Good day!

I currently have a Dell Precision Tower server running Windows 11 Pro. I have set up Hyper-V with four virtual machines (VMs), which will be used for business and client purposes.

One of the requirements is to allocate the server's NVIDIA RTX A6000 GPU across the four VMs. Ideally, I would like to allocate the GPU resources approximately as follows:

  • VM1: 20%
  • VM2: 20%
  • VM3: 30%
  • VM4: 30%

This would utilize 100% of the GPU resources.

I understand that GPU allocation/partitioning is primarily supported with Windows Server, and I have already tried using GPU-P on my current Windows 11 Pro setup. However, GPU-P is not working, and I am receiving an error/status message from the VMs.

Given this situation, could anyone recommend an alternative solution or configuration that would allow me to share the NVIDIA RTX A6000 GPU across the four Hyper-V VMs? If GPU-P is not supported for this particular setup, I would also appreciate any recommendations regarding other technologies or approaches that I could consider.

Any guidance or suggestions would be greatly appreciated.

Thank you in advance for your time and assistance!

Windows for business | Windows Client for IT Pros | Storage high availability | Virtualization and Hyper-V
0 comments

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. Hoang Le 4,500 Reputation points Independent Advisor
    2026-08-25T02:27:20.2466667+00:00

    Hello Ino Sone,

    On Windows 11 Pro, Hyper‑V does not support GPU partitioning (GPU‑P) for discrete GPUs like the NVIDIA RTX A6000. GPU‑P is only available in Windows Server Datacenter editions and requires specific drivers and host OS support. On client editions such as Windows 11 Pro, the only supported option is Discrete Device Assignment (DDA), which passes the entire GPU through to a single VM. This means you cannot split the GPU into percentages across multiple VMs in your current setup.

    If your requirement is to share one GPU across several VMs with defined resource allocation, you would need to move to Windows Server 2022 Datacenter with Hyper‑V and configure GPU‑P, provided the GPU driver stack supports it. Alternatively, NVIDIA offers vGPU (Virtual GPU) technology through their GRID licensing, which allows partitioning of RTX A6000 resources into profiles that can be assigned to multiple VMs. This requires NVIDIA vGPU software and licensing, plus a supported hypervisor such as VMware vSphere or Citrix Hypervisor. Hyper‑V itself does not natively support NVIDIA vGPU profiles.

    In short, on Windows 11 Pro you cannot achieve the 20/20/30/30 split you described. The practical paths are either upgrading to Windows Server Datacenter with GPU‑P support or adopting NVIDIA vGPU with the appropriate licensing and hypervisor. This ensures proper GPU sharing without unsupported workarounds.

    If my answer is useful for you, please hit Accept the answer to support me.

    Thank you,

    HL.

    Was this answer helpful?

    1 person found this answer helpful.
    0 comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.