AI News

Trending story

NVIDIA AICR adds AIBOM component based on k8s-aibom

1 reports1 story updatesUpdated Nov 12 01:00

Understand this story

AI overview

NVIDIA AI Cluster Runtime (AICR) has added a new AIBOM component, built on the open-source project k8s-aibom, which is used to track AI workloads running on Kubernetes and output CycloneDX 1.6 records with confidence grading and evidence localization. The two preset recipes h100-gke-cos-inference and h100-gke-cos-inference-dynamo enable this component by default, while other GKE recipes can opt in via --runtime-inventory. The integration was verified with AICR v1.2.0 pinning k8s-aibom 1.6.2, showing under steady-state observation approximately 4 API-server requests per minute, an average of 2.2 millicores of CPU, and 28 to 34 MiB of controller memory usage.

AI-generated from public reports

Latest development · Report

NVIDIA AICR integrates k8s-aibom to provide runtime AI asset inventory for GPU platforms

Developments

1 confirmed developments

  1. NVIDIA AICR adds k8s-aibom-based AIBOM component

    Source:NVIDIA Developer · Generative AI

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 12

  1. NVIDIA Developer · Generative AIFirst-party

    NVIDIA AICR integrates k8s-aibom to provide runtime AI asset inventory for GPU platforms

    NVIDIA AI Cluster Runtime (AICR) adds an AIBOM component based on the open-source project k8s-aibom, used to track AI workloads running on Kubernetes and output CycloneDX 1.6 records with confidence grading and evidence localization. The two preset recipes h100-gke-cos-inference and h100-gke-cos-inference-dynamo enable this component by default, while other GKE recipes can opt in via --runtime-inventory. The integration was validated with AICR v1.2.0 pinning k8s-aibom 1.6.2; under steady-state observation it makes about 4 API-server requests per minute, with an average of 2.2 millicores CPU and 28 to 34 MiB controller memory usage.

Related stories

  1. Related storiesGoogle launches Gemini 4 Hydrogen on Antigravity
  2. Related storiesMistral releases Large 4 weights on Hugging Face
  3. Related storiesNVIDIA releases OpenShell reference architecture
  4. Related storiesAmazon launches the TorchNeuron open-source toolkit