Install drivers for NVIDIA RTX Virtual Workstations (vWS)

Linux Windows

If you need GPUs for hardware accelerated graphics on a remote desktop orfor gaming, you can create virtual workstations that useNVIDIA RTX Virtual Workstations (vWS)(formerly known as NVIDIA GRID).

NVIDIA RTX Virtual Workstations are ideal for running workloads that useAPIs such as Vulkan, OpenGL, or Direct3D on a cloud remote desktop.

When you select aGPU that supports visualizationfor a virtual workstation, youcan add an NVIDIA RTX Virtual Workstation license to your VM. For moreinformation about pricing, seeGPU pricing.

To set up an NVIDIA RTX Virtual Workstation, you need to do thefollowing:

  1. Create a VM that has a visualization-capable GPU attached and virtualworkstation enabled.
  2. Install adriver for the virtual workstation.

After you create your virtual workstation, you can connect to it using a remotedesktop protocol such as Windows Remote Desktop Protocol (RDP),HP Anyware®, or VMware® Horizon View.

Before you begin

Recommended minimum NVIDIA virtual GPU software

NVIDIA RTX Virtual Workstations (vWS) are supported on the followingmachine series.

  • G4: you can enable NVIDIA RTX PRO 6000 vWS on these machine types.
  • G2: you can enable NVIDIA L4 vWS on these machine types.
  • N1: you can enable NVIDIA T4, V100, or P100 vWS GPUs on these machine types.

The following table provides the recommendedNVIDIA virtual GPU software software forthese machine types.

NVIDIA vGPU software versionSupported machine seriesRecommended NVIDIA driver (Linux)Recommended NVIDIA driver (Windows)vGPU software EOS date
19G4, G2, N1Latest R580 driverLatest R580 driverJuly 2028
18G2, N1Latest R570 driverLatest R570 driverMarch 2026
171G2, N1550.90.07 or later552.55 or laterApril 20251
16G2, N1535.183.01 or later538.67 or laterJuly 2026

1NVIDIA virtual GPU software version 17 has reached its end ofsupport. We recommend installing a later version of the virtual GPU software.

Limitations

For Ubuntu 24.04 NVIDIA RTX Virtual Workstations, the available NVIDIA GRID driverversions might not fully supportDirect Rendering Infrastructure 3 (DRI3).This lack of full support can affect GPU acceleration for certain desktopapplications and virtual desktop environments, such asVirtual Network Computing (VNC)andChromium. If youneed to run these applications, use Ubuntu 22.04.

Automated driver installation

You can use an open source script fromGitHubto automate the installation of RTX Virtual Workstation drivers.This is the same script used for regular driver installation. It automaticallydetects that your instance has RTX Virtual Workstation enabled and installs thecorrect driver.

Linux

Use these instructions to install GPU drivers on a running VM.

Supported operating systems

The Linux installation script was tested on the following operatingsystems:

  • Debian 12 and 13
  • Red Hat Enterprise Linux (RHEL) 8 and 9
  • Rocky Linux 8 and 9
  • Ubuntu 22 and 24

If you use this script on other operating systems, the installationmight fail. This script can install NVIDIA driver as well asCUDA Toolkit.

To install the GPU drivers and CUDA Toolkit, complete the following steps:

  1. Ensure that Python 3 is installed on your operating system.

  2. Download the installation script.

    curl -L https://storage.googleapis.com/compute-gpu-installation-us/installer/latest/cuda_installer.pyz --output cuda_installer.pyz

  3. Run the installation script.

    sudo python3 cuda_installer.pyz install_driver

    The script takes some time to run. It will restart your VM. When the VMrestarts, run the script again to continue the installation.

  4. Verify the installation. SeeVerify that the driver is installed.

  5. You can also use this tool to install the CUDA Toolkit. To install theCUDA Toolkit, run the following command:

    sudo python3 cuda_installer.pyz install_cuda

The script will take a while to run. It will restart your VM. When theVM restarts, run the script again to continue the installation.

  1. Verify the CUDA toolkit installation.

    python3 cuda_installer.pyz verify_cuda

  2. After you complete the installation, you must reboot the VM.

Linux (startup script)

Use these instructions to install GPU drivers during startup of a VM.

Supported operating systems

The Linux installation script was tested on the following operatingsystems:

  • Debian 12 and 13
  • Red Hat Enterprise Linux (RHEL) 8 and 9
  • Rocky Linux 8 and 9
  • Ubuntu 22 and 24

If you use this script on other operating systems, the installationmight fail. This script can install NVIDIA driver as well asCUDA Toolkit.

Use the followingstartup scriptto automate the driver and CUDA Toolkit installation:

#!/bin/bashiftest-f/opt/google/cuda-installerthenexitfimkdir-p/opt/google/cuda-installercd/opt/google/cuda-installer/||exitiftest-fcuda_installationthenexitficurl-fSsL-Ohttps://storage.googleapis.com/compute-gpu-installation-us/installer/latest/cuda_installer.pyzpython3cuda_installer.pyzinstall_cuda

Windows

This installation script can be used on GPU instances that have secure boot enabled.It supports Windows Server 2019 and later.

This script installs an NVIDIA RTX Virtual Workstation (vWS) compatibledriver. If you don't have a vWS license, vWS features won't be availableon your instance.

Open a PowerShell terminalas an administrator, then complete the followingsteps:

  1. Download the script.

    Invoke-WebRequest https://github.com/GoogleCloudPlatform/compute-gpu-installation/raw/main/windows/install_gpu_driver.ps1 -OutFile C:\install_gpu_driver.ps1

  2. Run the script.

    C:\install_gpu_driver.ps1

    The script takes some time to run. No command prompts are given during theinstallation process. Once the script exits, the driver is installed.

    This script installs the drivers in the following default location onyour VM:C:\Program Files\NVIDIA Corporation\.

  3. Verify the installation. SeeVerify the GPU driver install.

Manual driver installation

Complete the following steps on your VM.

  1. For Linux operating systems, install the build tools for your Linux version.

    CentOS/RHEL

    sudo yum group install "Development Tools"

    Debian/Ubuntu

    1. Install the build tools.

      sudo apt update
      sudo apt install -y build-essential
      sudo apt install -y libvulkan1

    2. If you are using Ubuntu 22.04, update thegcc version for the NVIDIAdriver.

      sudo apt install -y gcc-12
      sudo apt install -y linux-headers-$(uname -r)
      sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-12 12
      sudo update-alternatives --config gcc

    SLES

    • SLES 12

      sudo zypper install -t pattern devel_basis
      sudo zypper install gcc8

    • SLES 15

      sudo SUSEConnect -p sle-module-development-tools/15/x86_64
      sudo zypper install gcc8

  2. For Linux operating systems, install the Linux kernel headers.SeeVerify the System has the Correct Kernel Headers and Development Packages Installedin the NVIDIA documentation.

  3. For all operating systems, install drivers.

    Caution: The drivers provided in this section are for use withCompute Engine only. If you need drivers for other platforms, see theNVIDIA Driver Download page.

    Linux

    1. Download the driver. For a full list of NVIDIA RTX Virtual Workstationdrivers that you can use on Compute Engine, seeDrivers for the NVIDIA RTX Virtual Workstation.

      For example, to download the NVIDIA 580.126.09 driver, run the followingcommand:

      curl -Ohttps://storage.googleapis.com/nvidia-drivers-us-public/GRID/vGPU19.4/NVIDIA-Linux-x86_64-580.126.09-grid.run

    2. Start the installer.

      For example, to start the NVIDIA 580.126.09 driver, run the followingcommand:

      sudo bashNVIDIA-Linux-x86_64-580.126.09-grid.run

    3. During the installation, choose the following options:

      • If you are prompted to install 32-bit binaries, selectYes.
      • If you are prompted to modify thex.org file, selectNo.

    Windows Server

    1. Depending on your version of Windows Server, download therequired NVIDIA RTX Virtual Workstation drivers.

      For a full list of NVIDIA RTX Virtual Workstation driversthat you can use on Windows VMs that are running on Compute Engine,seeDrivers for the NVIDIA RTX Virtual Workstation.

    2. Run the installer, and choose theExpress installation.

    3. After the installation is complete, restart the VM. When you restart,you are disconnected from your session.

    4. Reconnect to your instance using RDP or a PCoIP client.

  4. This step is only required for Linux-based NVIDIA RTX Virtual Workstationwith attached T4 or L4 GPUs.

    If you installed an NVIDIA 510 or later driver, you need to disable theGSP firmware.SeeDisabling GSP Firmwarein the Nvidia documentation. When completing this step,keep the following in mind:

    • Usesudo to run the commands to create and update the configuration file.
    • To reboot the VM, you can usesudo reboot orstop and start the VM.

Verify that the driver is installed

Linux

After you install the drivers, verify that the driver is installedand that the NVIDIA RTX Virtual Workstation license is active.

  1. To verify the driver installation, run the following command:

    nvidia-smi

    The output of the command looks similar to the following:

    +-----------------------------------------------------------------------------------------+| NVIDIA-SMI 580.126.09             Driver Version: 580.126.09     CUDA Version: 13.1     |+-----------------------------------------+------------------------+----------------------+| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC || Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. ||                                         |                        |               MIG M. ||=========================================+========================+======================||   0  Tesla T4                       On  |   00000000:00:04.0 Off |                    0 || N/A   53C    P8             17W /   70W |       0MiB /  15360MiB |      0%      Default ||                                         |                        |                  N/A |+-----------------------------------------+------------------------+----------------------++-----------------------------------------------------------------------------------------+| Processes:                                                                              ||  GPU   GI   CI              PID   Type   Process name                        GPU Memory ||        ID   ID                                                               Usage      ||=========================================================================================||  No running processes found                                                             |+-----------------------------------------------------------------------------------------+

  2. To verify the NVIDIA RTX Virtual Workstation license, run thefollowing command:

    nvidia-smi -q

    To validate GPU acceleration for desktop applications, look for thevGPU Software Licensed Product section in the output. On a properlylicensed NVIDIA RTX Virtual Workstation VM, you should see the following:

    vGPU Software Licensed Product Product Name                      : NVIDIA RTX Virtual Workstation License Status                    : Licensed (Expiry: Permanent)

    This output indicates that the driver is functioning and able to provideaccelerated graphics.

    If you see a different output, the driver is not licensed forNVIDIA RTX Virtual Workstation and you won't get GPU acceleration.

    For example, if the VM wasn't created with the correct--accelerator-type that enables vWS, you might see the following output:

    vGPU Software Licensed Product Product Name                      : NVIDIA Virtual Applications License Status                    : Licensed (Expiry: N/A)

    This can happen if the VM was not created with the NVIDIA RTX VirtualWorkstation license enabled. To resolve this issue, you must recreate your VMusing a GPU model that supportsNVIDIA RTX Virtual Workstationand make sure to enable the virtual workstation when you attach the GPUs.For detailed instructions, review theCreate an instanceinstruction for your machine type.

Windows Server

  1. Connect to your Windows instance using RDP or a PCoIP client.

  2. Right-click the desktop, and selectNVIDIA Control Panel.

  3. In the NVIDIA Control Panel, from the Help menu, selectSystem Information. The information shows the GPU that the VM isusing, and the driver version.

What's next?

Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2026-02-12 UTC.