ai-cloud
Bare Metal GPU Provisioning for AI Clouds
Provision GPU servers for tenant workloads. vMetal orchestrates registered hardware through infrastructure drivers, and vCluster joins dedicated workers to customer clusters as Private Nodes.
Trusted by the fastest-growing AI cloud providers
Problem
GPU Provisioning Slows Revenue
AI cloud providers lose competitive ground when hardware sits idle and tenants wait.
Racks Sit Idle Too Long
Manual provisioning workflows delay time to first workload, costing revenue on every GPU hour your cluster is not serving tenants.
Machine and Cluster Workflows Diverge
Without an integrated stack, teams stitch together k3s, kubeadm, and custom scripts, then spend significant time maintaining what they assembled.
Customers Need More Than Servers
Selling bare metal GPUs alone is a race to the bottom. Customers want the cloud experience, not raw compute.
Solution
Zero Touch from GPU Rack to Production
vMetal orchestrates GPU server provisioning through configured infrastructure drivers. vCluster attaches the resulting worker capacity as Private Nodes to independent tenant clusters, keeping machine automation and tenant operations connected. vCluster Labs serves 50+ GPU clouds & Fortune 500s.
The Complete Bare Metal GPU Stack
From server registration to isolated tenant clusters, every layer is covered in one integrated platform.
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 GPU server onboarding. Network automation uses supported integrations such as Netris.
- Stable Machine API
- Bare metal and VM provisioning
- Driver-specific lifecycle automation
Kubernetes Layer
K8s Running Directly on Bare Metal
vCluster Standalone runs Kubernetes as a CNCF-certified binary on Linux, without an existing Kubernetes cluster. It supplies a Kubernetes foundation for GPU server onboarding; vMetal separately orchestrates machine and OS provisioning.
- CNCF-certified Kubernetes binary
- Runs directly on Linux
- No existing Kubernetes cluster required
Dynamic Scaling
Automatic GPU Node Provisioning
Karpenter-based Auto Nodes requests worker capacity according to configured policies. With the vMetal provider, Machine requests provision available infrastructure for GPU server onboarding; inventory, driver readiness and release rules determine actual capacity.
- Karpenter-based node management
- Supported infrastructure provider integration
- Inventory-aware capacity policies
Tenant Isolation
Tenant APIs Without Dedicated Servers
Each tenant has an independent Kubernetes API, data store and RBAC boundary. Lightweight control-plane hosting reduces dedicated server requirements for GPU server onboarding; 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
AI Workload Integration
Supported AI Workload Templates
Use supported workload templates for GPU server onboarding, 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
What is bare metal GPU provisioning and why does it matter for AI clouds?
Bare metal GPU provisioning is the process of taking physical GPU servers through PXE boot, OS installation, network setup, and Kubernetes registration. For AI cloud providers, every hour a rack sits unprovisioned is revenue lost. Zero-touch provisioning removes the manual steps, so teams go from powered-on hardware to production GPU compute without custom scripts.
How does vMetal handle GPU server lifecycle management?
vMetal exposes a stable Machine API above infrastructure-specific drivers for bare metal and VMs. Drivers handle supported provisioning mechanics such as imaging, registration and cleaning. Operators configure inventory, credentials and images before automation runs. vCluster can then use provisioned machines as dedicated Private Nodes, connecting the infrastructure lifecycle to the tenant cluster that needs the capacity.
Do I need an existing Kubernetes cluster to use vMetal?
vMetal requires vCluster Platform and a connected Kubernetes cluster for its Metal3/Ironic provisioning components. If you are starting without Kubernetes, vCluster Standalone can bootstrap that foundation as a binary on Linux. Once infrastructure drivers, hardware inventory, images and credentials are configured, vMetal can automate supported provisioning workflows and supply machines for tenant Private Nodes.
How quickly can an AI cloud provider go from bare metal to serving tenants?
Boost Run went from decision to production launch in under 45 days for its managed Kubernetes service. That is a customer example rather than a promised implementation time. Your launch depends on machine inventory, driver configuration, networking, storage, templates and service integrations. vMetal and vCluster connect machine orchestration with tenant cluster delivery, reducing the infrastructure workflows you need to build yourself.
How does bare metal GPU provisioning connect to tenant isolation?
vMetal handles the hardware provisioning layer while vCluster delivers isolated tenant clusters on the same bare metal. Each tenant gets its own control plane (API server, data store, and RBAC) running as a lightweight process. You provision GPU racks once and serve many tenants, each behind its own control plane and, with Private Nodes, on its own dedicated nodes, without buying a separate physical cluster per customer.
Is vMetal compatible with NVIDIA GPU infrastructure?
Private Nodes are worker nodes joined to one tenant cluster, with that tenant's own CNI and storage configuration. They can be physical machines or supported VMs. Use dedicated bare metal when the service requires exclusive physical servers. vMetal can provision the underlying machines, while platform policies define which capacity a tenant may request and how it is released.
Start Provisioning GPU Bare Metal Faster
See how vMetal goes from rack to production Kubernetes without the manual overhead.
