GPU - Get Started
Virtual machines equipped with NVIDIA Quadro GPUs. Ideal for complex processing and GPU-optimized workloads.
Deploy a GPU Linode Instance
Log in to the Cloud Manager with the username and password you created when signing up.
At the top of the page, click Create and select Linode.
NoteUse a StackScript to quickly deploy software platforms and system configuration options to your Linux distribution. Some of the most popular StackScripts do things like install a LAMP stack, VPN, or WordPress. You can read more about Stackscripts and how they work in our Automating Deployments with Stackscripts guide.
Choose the region where you would like your Linode to reside. If you’re not sure which to select, see our How to Choose a Data Center guide. You can also generate MTR reports for a deeper look at the route path between you and a data center in each specific region.
Select a GPU Linode plan.
Give your Linode a label. This is a name to help you easily identify it within the Cloud Manager’s Dashboard. If desired, assign a tag to the Linode in the Add Tags field.
Create a root password for your Linode in the Root Password field. This password must be provided when you log in to your Linode via SSH. The password must meet the complexity strength validation requirements for a strong password. Your root password can be used to perform any action on your server, so make it long, complex, and unique.
Click Create. You are directed back to the Linode’s page, and this page reports the status of your Linode as it boots up. You can now use the Cloud Manager to:
After the Linode GPU is online, install the appropriate NVIDIA drivers:
Install NVIDIA Driver Dependencies: Some dependencies need to be installed before the drivers are installed.
NVIDIA Driver CUDA Installation: If you’re running Ubuntu 18.04, CentOS 7, or OpenSUSE, install the NVIDIA driver with CUDA.
NVIDIA Driver Manual Installation: If you’re running Debian 9, install the NVIDIA driver manually.
This page was originally published on