New Latest release on GitHub

Quickstart

Try the whole ladder
without touching a GPU.

  1. 1
    Install KubeNeuron

    CRDs, operator, secrets, TLS bootstrap and a starter config in one run. Pinning the version selects the matching signed images and manifest.

  2. 2
    Post a synthetic XID 79

    The end-to-end test parks the incident at the reboot rung awaiting approval. No cluster, no hardware, about fifteen minutes.

  3. 3
    Approve it, read the audit trail

    Approve over the API and assert the full trail through to resolution. Then decide which rungs deserve to act on their own.

  • Kubernetes 1.29+ for the operator path
  • NVIDIA GPU Operator supplies dcgm-exporter
  • VictoriaMetrics operator for the metrics path
  • Go 1.25+ to build from source
# CRDs, operator, secrets, TLS bootstrap, starter config — one run
# pin the version: it selects the matching signed images and manifest
curl -sfL https://github.com/kubeneuron/kubeneuron/releases/latest/download/install.sh \
  | bash -s -- --version latest

# ✓ KubeNeuron is Ready. → prints your panel sign-in and API token
kubectl -n kube-neuron port-forward service/kubeneuron-controller 8080:8080

Build the safe path first

Run a calmer GPU fleet.

Read the source, run the dry-run ladder, and tell us where the policy model breaks.