One-click Kubernetes clusters on your own infrastructure, powered by Cluster API. All the convenience of managed K8s — none of the cloud lock-in.
Kubernetes is powerful. It's also a full-time job. Between bootstrapping clusters, managing upgrades, configuring networking, and praying to the etcd gods — running K8s on-prem has always been a project, not a feature. We think it should be a checkbox.
flex.plane uses Cluster API (CAPI) to provision and manage Kubernetes clusters on your Proxmox infrastructure. CAPI is the Kubernetes-native standard for cluster lifecycle management — but we've wrapped it so you never have to write a single CRD manifest. Click a button, pick a size, get a cluster.
Deploy production-ready Kubernetes clusters directly on your Proxmox infrastructure. No manual kubeadm, no YAML maze.
Built on CAPI — the Kubernetes-native way to manage clusters. Declarative, reconcilable, and well-understood by the ecosystem.
Scale worker nodes up and down based on demand. Define min/max boundaries and let the cluster autoscaler handle the rest.
Upgrade Kubernetes versions with zero downtime. Nodes are drained, replaced, and rejoined automatically.
Clusters are automatically connected to your Tailscale mesh. Pod networking, service discovery, and ingress work out of the box.
Kubernetes clusters inherit VDC-level access controls. Tenant isolation without manual RBAC configuration.
Automatic provisioning of persistent volumes backed by Proxmox storage. CSI driver included, no manual setup.
Clusters come pre-configured for GitOps workflows. Point Flux or ArgoCD at your repo and you're deploying.
Pull images from your private registries without cluster-level credential management headaches.
Define reusable cluster templates — control plane size, worker flavors, networking CNI, add-ons. Stamp out consistent clusters every time.
We're excited to announce that we're only a few weeks away from sending out the first invitations to our early access program. Enter your E-Mail to be notified once it drops. And don't worry, we hate spam as much as you.