This article will cover installing and setting up official Nvidia drivers on Red Hat Enterprise Linux; this method has been tried and tested on RHEL 8.

It works best on the default kernel that ships with RHEL 8. If you have a custom Kernel or modified setup, you might need extra steps. The additional steps are out of the scope of this article.

Let’s get straight into it. (The step speed may vary depending on your internet connection)

Step 1. Adding the official Nvidia drivers repository to our package managers repository list.

sudo dnf config-manager --add-repo language: JavaScript (javascript)

Step 2. Installing the kernel-devel and headers used by the drivers.

sudo dnf install kernel-devel-$(uname -r) kernel-headers-$(uname -r)Code language: JavaScript (javascript)

Step 3. Installing Nvidia drivers and settings application.

The command below will download the latest stable driver available.

sudo dnf install nvidia-driver nvidia-settings

Step 4. Installing CUDA drivers. (This is optional but recommended)

sudo dnf install cuda-driver

Step 5. Reboot your system.

Step 6. Verify that it is working by running nvidia-smi


The nvidia-smi command should display information about your graphics card; if this command fails to show the details or throws an error after the command is entered, the driver did not install it correctly.

To remove the drivers altogether.

Run the command below.

sudo dnf remove nvidia-driver nvidia-settings cuda-driver kernel-devel-$(uname -r) kernel-headers-$(uname -r)Code language: JavaScript (javascript)

Running RHEL 9 follow this guide

I compiled a list of software and services that I use to improve my workflow, here is the link to the list.

Darryl Dias

I’m Darryl. I’m a 3D Artist, Programmer and Linux enthusiast. On this site I share my insights, tips and tricks, tutorials, methods and best practices.

Join the Conversation


  1. Thanks Darryl! IMHO, this is the most sustainable approach, especially when dealing /w UEFI devices: all other sources I stumbled upon recommend manual download, specifying the key files etc. – especially bad when you landed at the console after an upgrade.

    Worked like a charm for me!

  2. i have the drive installed – however not used by gnome desktop – how to fix:
    on redhat 8:

    Tue Dec 13 13:05:04 2022
    | NVIDIA-SMI 510.47.03 Driver Version: 510.47.03 CUDA Version: 11.6 |
    | 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 GRID P6-8Q On | 00000000:02:00.0 Off | 0 |
    | N/A N/A P8 N/A / N/A | 13MiB / 8192MiB | 0% Default |
    | | | N/A |

    | Processes: |
    | GPU GI CI PID Type Process name GPU Memory |
    | ID ID Usage |
    | 0 N/A N/A 28988 G /usr/libexec/Xorg 13MiB |

    ,,,,,,,,,,,,,,, but in gnome it is using this driver:

    llvmpipe (llvm 14.0.6, 256 bits)

    1. Hi J, You need to blacklist nouveau drivers, and also, if you can disable secure boot in your BIOS. Nvidia drivers run better with secure boot enabled.

Leave a comment

Your email address will not be published. Required fields are marked *