If you do not select to automatically install a GPU driver when you create an instance, you must manually install the GPU driver after the instance is created. This topic describes how to manually install a GPU driver on a GPU-accelerated instance.

Background information

You can install only GRID drivers on vgn6i or vgn5i instances. Therefore, the procedure in this topic does not apply to vgn6i or vgn5i instances. For more information about how to install GRID drivers, see Create an NVIDIA GPU-accelerated instance and Install an NVIDIA GRID driver on a GPU-accelerated virtualization Linux instance.


  1. Go to the NVIDIA official website.
  2. Manually search for the suitable driver.
    1. Select a product type, series, and product based on the GPU with which your instance type is equipped.
      The following table lists the information of GPUs with which different instance types are equipped.
      Item gn4 gn5 gn5i gn6v gn6i
      Product type Tesla Tesla Tesla Tesla Tesla
      Product series M-Class P-Series P-Series V-Series T-Series
      Product M40 Tesla P100 Tesla P4 Tesla V100 Tesla T4
    2. Select an operating system based on the image of the instance.
      If your instance runs Debian, select Linux 64-bit. If the drop-down list does not contain the operating system that you instance uses, click Show All Operating Systems in the lower part of the drop-down list.
    3. Select a CUDA Toolkit version.
    4. Select a language.
    5. Click SEARCH.
  3. Confirm the driver information, and click DOWNLOAD.
    If your instance runs a Linux operating system, do not close the download page. When you install the GPU driver, you may need to refer to the installation steps in the ADDITIONAL INFORMATION section.
  4. Install the GPU driver.
    • Install the GPU driver on a Linux instance:
      1. Download and install the kernel-devel and kernel-header packages based on your kernel version.
        Note The mismatch between kernel and kernel-devel versions results in a driver compilation error when the driver is installed from the .rpm file. You can run the sudo rpm –qa | grep kernel command on an instance to check whether the versions match. Make sure that the versions match and re-install the driver.
      2. Run the sudo rpm -qa | grep $(uname -r) command to check whether the kernel-devel and kernel-header packages are downloaded and installed.
        If information similar to the following content is displayed, the kernel-devel and kernel-header packages are installed. CentOS 7.3 is used in this example.
      3. Perform the following steps in the ADDITIONAL INFORMATION section on the download page to install the GPU driver.
        The steps in the ADDITIONAL INFORMATION section are shown in the following figure. Linux Ubuntu 14.04 64-bit is used in this example.nvdia-driver-en.png
    • Install the GPU driver on a Windows instance:
      Double-click the package and follow the prompts to complete the installation.
      Note On Windows instances where the installed GPU drivers have taken effect, Windows Remote Desktop Protocol (RDP) may not support DirectX- and OpenGL-based applications. In this case, you must install the Virtual Network Computing (VNC) service and client or use other protocols that support these applications, such as PC over IP (PCoIP) and XenDesktop HDX 3D.