Clusters

Updated on 21 April, 2026

Manage cluster subscriptions on Vultr to organize compute resources with shared networking and automated fabric provisioning for GPU and CPU workloads.


Add existing instances to a Vultr cluster in bulk or individually. GPU clusters automatically configure fabric networking when instances join.
Check available cluster types and plans for a specific Vultr region before creating a cluster. Filter results by VPS, bare-metal, or GPU fabric type.
Create a Vultr cluster using a reusable instance template that defines the plan, OS, SSH keys, and startup script for consistent deployments.
Create a Vultr cluster subscription by specifying a region, plan, and pool size using the Vultr API for VPS, bare-metal, or GPU fabric workloads.
Delete a Vultr cluster subscription to remove the cluster container and fabric associations while keeping the underlying instances on your account.
Retrieve all cluster subscriptions associated with your Vultr account using the API with cursor-based pagination support for large result sets.
Remove instances from a Vultr cluster in bulk or individually. Detached instances remain on your account as standalone servers with fabric removed.
Retrieve full details of a Vultr cluster including its label, region, type, pool configuration, attached instances, and current status via the API.
Retrieve GPU and fabric performance metrics for all instances in a Vultr cluster with configurable time periods of 1, 7, or 30 days via the API.
Update a Vultr cluster's label, hostname, and pool count to scale the cluster up or down by adjusting the desired and minimum pool size via the API.