NVIDIA Kubernetes Launch Kit
NVIDIA Kubernetes Launch Kit (l8k) is a CLI tool that automates the deployment of NVIDIA cloud-native networking on Kubernetes. It discovers cluster hardware, selects a deployment profile, generates Kubernetes YAML manifests, and optionally deploys them. Launch Kit supports SR-IOV, RDMA, InfiniBand, Host Device, MacVLAN, and Spectrum-X configurations, including heterogeneous clusters with mixed NIC and GPU hardware.
Note
Who this is for: operators and platform engineers deploying NVIDIA networking on a Kubernetes cluster. Launch Kit replaces hand-written NicClusterPolicy, NicNodePolicy, and SriovNetworkNodePolicy manifests with a discover-driven workflow.
Install
Install the l8k binary with the install script:
curl -fsSL https://raw.githubusercontent.com/nvidia/k8s-launch-kit/main/scripts/install.sh | sh
Or with Homebrew:
brew tap nvidia/l8k https://github.com/nvidia/k8s-launch-kit
brew install l8k
Where to Start
I want to… |
Go to |
|---|---|
See all installation options (container, build-from-source, version pinning) |
|
Get the mental model and core vocabulary |
|
Run an end-to-end walkthrough on a fresh cluster |
|
Pick the right profile for my cluster |
|
Run a specific phase — discover, generate, or deploy |
|
Configure heterogeneous clusters, presets, CI/CD, AI agents, or troubleshoot |
|
Look up flags or the config schema |