Nebius on Northflank

You can integrate your Nebius account to create and manage clusters using Northflank.

Navigate to Cloud → Provider links → Create provider link and select Nebius as the provider. Or use the direct link:

First, you will need to set up a service account with an authorized key on Nebius. The service account will need to be in the editors group for each project that should be managed by Northflank. Follow Nebius docs to:

• Prepare a service account
• Prepare an authorized key

To create a provider link, you need to supply the service account ID, the authorized key ID, and the private key PEM.

You can now configure and deploy new clusters in your Nebius account.

Note: You can edit the integration at any time to update the credentials, if required.

To successfully deploy a cluster on Nebius using Northflank, you must have the required resources available to your account.

You can view your account quota and request new quota from the Nebius console under Administration → Limits. In order to provision a cluster, you will require a cluster resource and the appropriate compute and virtual network resources.

Learn more on the Nebius documentation on quotas .

Navigate to Cloud → Clusters → Create cluster and select Nebius as the provider. Or use the direct link:

Enter a name for the cluster or generate one randomly. Choose your integration credentials. Select the project and a subnet within that project to deploy the cluster in.

Node pool configuration

You can now configure the node pools for your cluster. Node pools can also be added, deleted, and updated after creating your cluster. Click add node pool to add another pool. Node types which are not available in your chosen project will be grayed out in the selection box.

Some GPU node types can optionally be attached to a Nebius GPU cluster (an InfiniBand interconnect). In order to do this, you will first need to create the GPU cluster in your Nebius project, and then select it from the drop-down box.

GPU node types support spot instances, while CPU node types do not.

Each node can schedule up to 256 pods (minus system pods). The actual number of pods per node will usually be limited by resource requests and request modifiers for smaller nodes.

See deploy and scale node pools for more information on configuring nodes and node pools.

Configure advanced options

After adding your initial node pools, you can configure advanced options for the cluster, such as build infrastructure and resource request modifiers.

When you create the cluster, Northflank will begin installing system components in the node pools according to their capacity. Provisioning a node pool can take up to 20 minutes.