NVIDIA Open-Sources NVKE and NVKR for Managed Kubernetes on GPU Infrastructure

Executive Briefing
“NVIDIA has released two open-source projects named NVIDIA Kubernetes Engine and NVIDIA...”
Key Takeaways
- 1.NVIDIA released two open-source projects, NVIDIA Kubernetes Engine and NVIDIA Kubernetes Recipes, to run managed Kubernetes on GPU infrastructure.
- 2.NVIDIA Kubernetes Engine serves as the control plane for a managed Kubernetes service, while NVIDIA Kubernetes Recipes builds the OS images, container images, and packages those clusters depend on.
- 3.Both projects ship under Apache 2.0, are conformance tested on Kubernetes 1.36, and use SPIFFE/SPIRE attestation with TPM for node identity.
NVIDIA has released two open-source projects named NVIDIA Kubernetes Engine and NVIDIA Kubernetes Recipes to manage Kubernetes on GPU infrastructure. Both projects are currently in public preview under the Apache 2.0 license.
Key Developments and Technical Context
NVIDIA released two open-source projects, NVIDIA Kubernetes Engine and NVIDIA Kubernetes Recipes, to run managed Kubernetes on GPU infrastructure.
NVIDIA Kubernetes Engine serves as the control plane for a managed Kubernetes service, while NVIDIA Kubernetes Recipes builds the OS images, container images, and packages those clusters depend on.
Both projects ship under Apache 2.0, are conformance tested on Kubernetes 1.36, and use SPIFFE/SPIRE attestation with TPM for node identity.
Strategic Analysis and Industry Implications
Integrating artifact building and cluster management into open-source software helps streamline environment standardization for large-scale AI factories.
Outlook and Key Factors to Watch
Actual performance and provisioning-time benchmarks have not been published as the projects remain in public preview.







