ai-cloud
Provisioning Orchestration for AI Clouds
vMetal places one stable Machine API above interchangeable physical and virtual infrastructure drivers. vCluster uses that capacity for Private Nodes and managed tenant cluster products.
Trusted by the fastest-growing AI cloud providers
Problem
Why Manual Provisioning Breaks at Scale
AI cloud infrastructure becomes difficult to automate when every provisioning stack exposes a different API and lifecycle.
Slow, Fragmented Provisioning Stacks
Disconnected provisioning systems create separate workflows for bare metal, VMs, networking, and cluster attachment.
No Stable API Across Drivers
Direct dependencies on driver-specific APIs make infrastructure changes ripple into customer-facing automation.
GPU Lifecycle Left to Manual Ops
Manual Machine lifecycle work delays provisioning, release, cleaning, and capacity reuse.
Solution
One API to Orchestrate Every Machine Layer
vMetal provides one stable Machine API above infrastructure-specific drivers. Higher layers request capacity consistently while drivers keep ownership of their provisioning mechanics.
Built for GPU Rack to Production Orchestration
vMetal unifies physical and virtual Machine provisioning behind one API, while vCluster turns that capacity into tenant cluster products.
Bare Metal Orchestration
Automated Machine Provisioning
vMetal sends a Machine request to the selected driver and tracks the result through a consistent lifecycle.
- PXE boot to production automatically
- Bare Metal Machines as sellable inventory
- Full decommission lifecycle managed
Dynamic Scaling
Capacity Requests Through Auto Nodes
Auto Nodes can create and delete Machine requests to maintain Private Node capacity for tenant clusters.
- GPU nodes provisioned on workload trigger
- No manual scale-out operations
- Stable Machine API
Network Automation
Network Configuration by Driver
Network configuration follows the selected driver. Metal3 can integrate with Netris for hardware-backed L2 isolation when configured.
- VLANs and VXLANs automated per tenant
- Network automation through supported integrations layer
- Hard network isolation from day one
Tenant Cluster Delivery
Automatic Tenant Cluster Attachment
After provisioning, a Machine requested for tenant capacity can join its target cluster automatically as a standard Kubernetes node.
- Automatic cluster attachment
- Own API server and etcd per tenant
- No separate physical cluster required
Tenant Isolation
Private Nodes for Production Tenants
vCluster uses provisioned Machines as Private Nodes, assigning dedicated worker capacity to one production tenant cluster at a time.
- Private Nodes per tenant, the production default
- Per-tenant CNI and storage assigned
- No cross-tenant workload exposure
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 makes vMetal a provisioning orchestrator rather than a provisioning tool?
vMetal is the orchestration and lifecycle layer above infrastructure-specific drivers. The stable Machine API lets higher layers request capacity without depending on the mechanics of each provisioning stack.
What does machine provisioning orchestration cover end to end?
Every Machine moves through the same driver-neutral milestones: requested, scheduled, provisioned, joined when used by a tenant cluster, ready, and destroyed. Driver-specific systems perform the underlying infrastructure work.
How does vMetal integrate with Kubernetes cluster orchestration?
vMetal provisions the Machine. vCluster requests that capacity through Private Nodes and Auto Nodes, then the provisioned server joins the target tenant cluster as a standard Kubernetes node.
Can vMetal work with our existing provisioning infrastructure?
Yes, when a documented vMetal driver supports the infrastructure. Current drivers cover Metal3 bare metal, KubeVirt VMs, BCM existing installations, Terraform or OpenTofu, and Cluster API mechanisms.
How does Auto Nodes relate to machine provisioning orchestration?
Auto Nodes is one vCluster workflow that consumes the vMetal Machine API. It maintains configured Private Node capacity by creating and deleting Machine requests.
What GPU cloud providers use vMetal for provisioning orchestration?
vCluster powers 100K GPUs across 50+ GPU Clouds & Fortune 500s. Specific driver support and deployment references should be checked against the current vMetal documentation.
Turn Your GPU Racks Into a Cloud Product
See how vMetal orchestrates your full machine provisioning lifecycle end to end.
