---
title: "AICR v1.0: Open, stable, and verifiable GPU cluster configuration"
description: "GPU-accelerated Kubernetes clusters depend on compatible versions across host kernels, GPU drivers, container runtimes, networking, storage, operators, and workload frameworks, each on its own release cycle. A configuration that works for one service, GPU generation, and Kubernetes release may silently fail for another, and tracing version conflicts after deployment is slow. NVIDIA AI Cluster Runtime (AICR) v1.0 addresses this with version-locked, validated recipes that pin component combinations, render deployment artifacts for Helm, Argo CD, Flux, or Helmfile, and carry signed validation evidence from tested hardware. The release establishes a stable compatibility contract across its CLI, REST API, Go SDK, bundle layout, and artifact schemas, with merge-blocking checks on public interfaces. AICR now has over 100 contributors, with almost half from outside NVIDIA, and integrations from Pulumi Labs and Mirantis demonstrate consuming the same configuration through different tools."
---

# AICR v1.0: Open, stable, and verifiable GPU cluster configuration

[NVIDIA](https://yomu.fyi/company/nvidia-developer-blog) · Elizabeth Goodman · Oct 6, 2026

**Type:** Announcement

## Summary

GPU-accelerated Kubernetes clusters depend on compatible versions across host kernels, GPU drivers, container runtimes, networking, storage, operators, and workload frameworks, each on its own release cycle. A configuration that works for one service, GPU generation, and Kubernetes release may silently fail for another, and tracing version conflicts after deployment is slow. NVIDIA AI Cluster Runtime (AICR) v1.0 addresses this with version-locked, validated recipes that pin component combinations, render deployment artifacts for Helm, Argo CD, Flux, or Helmfile, and carry signed validation evidence from tested hardware. The release establishes a stable compatibility contract across its CLI, REST API, Go SDK, bundle layout, and artifact schemas, with merge-blocking checks on public interfaces. AICR now has over 100 contributors, with almost half from outside NVIDIA, and integrations from Pulumi Labs and Mirantis demonstrate consuming the same configuration through different tools.

## Context

GPU-accelerated Kubernetes clusters depend on compatible versions across many components, each on its own release cycle. A configuration that works for one service, GPU generation, and Kubernetes release may silently fail for another, and tracing version conflicts after deployment is slow and error-prone.

## Approach / What changed

AICR provides version-locked, validated recipes that pin component combinations, render deployment artifacts for Helm, Argo CD, Flux, or Helmfile, and carry signed validation evidence from tested hardware. It also establishes a stable compatibility contract across its CLI, REST API, Go SDK, bundle layout, and artifact schemas.

## Takeaways

- AICR v1.0 defines compatibility rules for its public CLI, REST API, Go SDK, bundle layout, and artifact schemas, with merge-blocking checks before changes merge.
- AICR provides four core capabilities: Snapshot (observed state), Recipe (desired configuration), Bundle (rendered artifacts), and Validation (comparison and checks).
- AICR has over 100 contributors, with almost half from outside NVIDIA, and integrations from Pulumi Labs and Mirantis.

## FAQ

### How does AICR ensure configuration compatibility?

AICR uses version-locked recipes that pin component combinations and carry signed validation evidence from tested hardware. The v1.0 release establishes a stable compatibility contract across its public interfaces with merge-blocking checks.

### How can operators validate a cluster against a recipe?

Operators can use the aicr CLI to validate the running cluster against a recipe, and where evidence is published, run the aicr evidence verify command to inspect signed results.

### How can contributors propose new recipes?

Contributors can propose recipes for environments not yet covered, validate them on their own clusters, and submit signed evidence for maintainer review.

**Tags:** [AI](https://yomu.fyi/topic/ai), [GPU](https://yomu.fyi/topic/gpu), [Kubernetes](https://yomu.fyi/topic/kubernetes), [NVIDIA](https://yomu.fyi/topic/nvidia)

- Source: [NVIDIA](https://developer.nvidia.com/blog/aicr-v1-0-open-stable-and-verifiable-gpu-cluster-configuration)
- Source URL: https://developer.nvidia.com/blog/aicr-v1-0-open-stable-and-verifiable-gpu-cluster-configuration
- Ingested by Yomu: 2026-10-06T18:00:13.802Z

[Read original post](https://developer.nvidia.com/blog/aicr-v1-0-open-stable-and-verifiable-gpu-cluster-configuration)