Installing the Linux GPU driver

Check your VM's configuration type:
1. Go to your VK Cloud personal account.
2. Go to Cloud Computing → Virtual Machines.
3. In the list, find the VM for which you need to install the GPU driver. Make sure the Type column shows a dedicated GPU configuration flavor.
Connect to the VM with the GPU, or create a new one if you haven't done so already.
Check whether the GPU driver is installed:
```
/usr/bin/nvidia-smi --query-gpu=driver_version --format=csv,noheader| head -n1
```
The following results are possible:
- A No such file or directory error means the driver is not installed.
- The command output shows a driver version. This means the driver is already installed and the NVIDIA stack was initialized successfully. Go to Updating the Linux GPU driver.
- nvidia-smi runs, but the output shows a ... couldn't communicate with the NVIDIA driver error. This means the driver is damaged or an incorrect version is used (for example, for vGPU). Reinstall the driver: first remove the GPU driver, then install it again by repeating the steps in this instruction.

Install the driver:

Ubuntu/Debian

Red Hat Enterprise Linux

Determine your distribution version:
```
cat /etc/os-release
```

Download and install the cuda-keyring_1.1-1_all.deb package from the NVIDIA CUDA repository for the corresponding version of your distribution:


wget https://developer.download.nvidia.com/compute/cuda/repos/<DISTRIBUTION><VERSION>/x86_64/cuda-keyring_1.1-1_all.deb && \sudo dpkg -i cuda-keyring_1.1-1_all.deb

Here:

<DISTRIBUTION> — the distribution name in Latin letters.
<VERSION> — the distribution version. Numbers only, without dots.

Examples: debian13, ubuntu2404.

Update the package manager indexes:
```
sudo apt update
```
Make sure the required or a newer driver version is available for installation:
```
apt-cache policy cuda-drivers
```
If the command succeeds, the output will include a Candidate field indicating the version to be installed from the cuda-drivers meta-package.

Install the driver:


sudo apt -y install cuda-drivers-<BRANCH_VERSION>

Here <BRANCH_VERSION> is the major version number of the driver you need to install. For example, 550.

Check that the driver is installed:


nvidia-smi --query-gpu=driver_version --format=csv,noheader| head -n1

If running the command results in a command not found error, add nvidia-smi to the environment variables manually:


echo 'export PATH="$PATH:/usr/bin"' | sudo tee /etc/profile.d/nvidia-smi-path.sh >/dev/null &&source /etc/profile.d/nvidia-smi-path.sh

Check again that nvidia-smi works:


nvidia-smi --query-gpu=driver_version --format=csv,noheader| head -n1

Reboot the VM.

Was this article helpful?