ai-cloud
Private GPU Cloud Kubernetes for AI Providers
Build an internal private GPU cloud for enterprise teams. vMetal supplies machines, and vCluster separates team Kubernetes environments with dedicated workers through Private Nodes.
Trusted by the fastest-growing AI cloud providers
Problem
The GPU Cloud Kubernetes Problem
Plain Kubernetes makes you choose between a strong tenant boundary and running the platform efficiently.
Namespace Isolation Is Too Weak
With namespaces, workloads can see cluster-wide agents and other tenants' nodes that they have no business seeing.
Separate Cluster Stacks Add Work
A full cluster per customer multiplies hardware, operational overhead and time to revenue with every new contract.
DIY Platform Takes Too Long
Building private GPU cloud tooling in-house takes a dedicated team and a long runway before the first customer is live.
Solution
Tenant Clusters Without Physical Cluster Overhead
vMetal provisions your private GPU infrastructure. vCluster gives enterprise teams independent Kubernetes tenant environments with dedicated Private Nodes, while Platform centralizes access, templates and fleet operations. vCluster Labs software powers 100K+ GPUs.
Built for Private GPU Cloud Kubernetes
One stack from bare metal provisioning to tenant clusters to workload isolation, built for GPU cloud providers.
Tenant Isolation
Isolated Tenant Clusters as Lightweight Pods
Each tenant has an independent Kubernetes API, data store and RBAC boundary. Lightweight control-plane hosting reduces dedicated server requirements for internal private GPU clouds; Private Nodes keep production workers tenant-specific.
- Independent tenant API and data store
- Tenant-scoped scheduling on Private Nodes
- No dedicated control-plane servers per tenant
Bare Metal
Zero-Touch GPU Server Provisioning
vMetal exposes one Machine API above bare metal and VM provisioning drivers. Registered inventory, images and credentials enable repeatable machine lifecycle workflows for internal private GPU clouds. Network automation uses supported integrations such as Netris.
- Stable Machine API
- Bare metal and VM provisioning
- Driver-specific lifecycle automation
Worker Isolation
Private Nodes Per Tenant
Private Nodes dedicate worker capacity to one tenant cluster, with its own CNI and storage configuration. For internal private GPU clouds, select physical servers or VMs according to the required infrastructure boundary.
- Dedicated worker nodes per tenant
- Tenant-owned CNI and storage configuration
- Physical servers or supported VMs
Workload Security
Kernel-Native Workload Isolation
vNode uses Linux user namespaces and seccomp to restrict workload privileges and system calls. It adds runtime hardening for internal private GPU clouds without a guest kernel or hypervisor, complementing dedicated workers rather than allocating GPUs.
- Linux user namespaces and seccomp
- No guest kernel or hypervisor
- Helps limit workload breakout
AI Workload Integration
Supported AI Workload Templates
Use supported workload templates for internal private GPU clouds, including the NVIDIA Run:AI partner integration. Configure GPU, network and storage prerequisites, and validate the selected template for Private Nodes before offering it to production customers.
- Native NVIDIA Run:AI catalog
- Validate Private Node configuration
- Release-specific integration prerequisites
Why vCluster
This isn’t a side project. Behind every vCluster deployment is 5+ years of deep K8s engineering, security hardening, and battle-tested infrastructure work at massive scale.
100K+
GPUs Powered
50+
GPU Clouds & F500s
<45
Days to Launch
30K
GitHub Stars
Get Started in 3 Steps
1
Schedule a Demo
Talk to our team about your stack
2
Deploy vCluster
Deploy vCluster on your infra in minutes
3
Onboard Your Tenants
Go live with a hyperscaler-grade tenant experience in days
FAQs
How does vCluster simplify an internal enterprise GPU cloud?
vCluster runs an independent Kubernetes control plane for each enterprise team without dedicated control-plane servers per team. Private Nodes provide dedicated workers, while Platform centralizes templates, access and operations. vMetal orchestrates the machine layer. This separates team administration from infrastructure management while letting your enterprise select supported networking, storage and runtime controls for each environment.
How does vCluster separate enterprise teams’ GPU workloads?
Private Nodes are the recommended production model: each customer gets dedicated worker nodes with separate CNI and CSI. Your own internal teams can run dev, test and CI clusters on shared nodes with resource quotas. vNode adds a workload boundary using Linux user namespaces and seccomp, without a hypervisor, and helps prevent container breakout.
Can vCluster deploy directly on bare metal GPU servers?
vCluster Standalone runs as a Kubernetes binary directly on Linux without requiring another Kubernetes cluster, k3s or kubeadm. It can provide an initial Kubernetes foundation. vMetal is a separate infrastructure orchestration layer that requires vCluster Platform and a connected Kubernetes cluster for its Metal3/Ironic path. Hardware registration, images, credentials and driver setup remain prerequisites for machine provisioning.
What determines an enterprise private GPU cloud launch timeline?
An enterprise AI factory launch depends on GPU inventory, provisioning drivers, networking, storage, identity and the workload environments teams need. vMetal orchestrates the machine layer, while vCluster provides tenant environments and self-service templates. Validate capacity, application readiness and operational ownership before rollout. The NVIDIA DGX reference architecture is the approved supporting asset for this audience.
Is vCluster Kubernetes conformant for enterprise GPU cloud customers?
vCluster is a CNCF-certified Kubernetes distribution and exposes standard Kubernetes APIs in tenant environments. This supports familiar clients, Helm charts and cluster-scoped resources such as CRDs. Certification covers Kubernetes conformance rather than an entire AI platform or customer deployment. Validate the selected version, networking, storage, GPU drivers and runtime integrations for the workloads you plan to support.
What supports evaluating vCluster for an enterprise GPU deployment?
The approved supporting asset for enterprise AI factory deployments is the vCluster on NVIDIA DGX reference architecture. It describes a deployment approach rather than a customer-specific rollout result. Evaluate tenant boundaries, dedicated worker capacity, networking, storage and workload integrations against your enterprise requirements. AI cloud customer launch figures should not be treated as enterprise AI factory outcomes.
Launch Your Private GPU Cloud Kubernetes
See how GPU cloud providers give every customer a tenant cluster on Private Nodes.
