Nvidia AI Cluster Runtime packages Dynamo into tested Kubernetes recipes
Nvidia has released NVIDIA AI Cluster Runtime (AICR), version-locked recipes that bundle its Dynamo inference framework with Kubernetes dependencies to streamline deployment.
Nvidia wants to take the guesswork out of running its Dynamo inference framework on Kubernetes. Its tool for the job is NVIDIA AI Cluster Runtime (AICR), which bundles Dynamo and its Kubernetes dependencies into tested recipes that teams can deploy and then verify.
AICR applies the idea of version-locked recipes to GPU clusters, meaning every component is pinned to a specific release that Nvidia has confirmed works with the others. That matters because GPU-accelerated Kubernetes clusters have many moving parts where a single mismatched version can stall an entire deployment.
Workflow and Validation
The runtime organizes deployment into a sequence of workflows: snapshot, recipe generation, bundle rendering, deployment, and validation. The snapshot step captures the current state of a cluster, so teams know exactly what they started with before changing anything.
Nvidia says AICR supports multiple cloud services and accelerators, backed by signed validation evidence. Put plainly, teams get a cryptographically signed record showing a given configuration was tested, not just a promise that it probably works.
Version Requirements
AICR hit version 1.0 in early October 2026, marking its first stable release. Documentation for deploying Dynamo through the runtime followed around October 30 to 31, 2026, including guidance for Amazon's Elastic Kubernetes Service (EKS) running on H100 and GB200-class GPUs.
Teams that want to use Dynamic Resource Allocation (DRA) need Kubernetes version 1.34 or later, as DRA is a Kubernetes feature for handing out specialized hardware like GPUs more flexibly.
Dynamo Framework Benefits
Dynamo is Nvidia's open-source framework for distributed inference, where a trained model actually answers questions, generates text, or produces images for users. It's built for disaggregated serving, splitting work across GPU workers, and offers KV-cache-aware routing to avoid redundant processing.
For teams running AI in production, AICR's snapshot and validation steps target configuration drift directly, providing an audit trail useful for debugging and compliance. The signed validation evidence offers documentation that configurations were actually tested, not just asserted to work.
Related
