FreeLLMNVIDIA launches open‑source DSX OS to certify and run GPU‑Kubernetes clusters
The DSX AI Factory Platform adds tools for configuring, certifying and operating accelerated Kubernetes clusters.
NVIDIA announced the DSX AI Factory Platform, which includes an open‑source software layer called DSX OS.
DSX OS bundles three projects: AI Cluster Runtime (AICR) for reproducible configuration, Cluster Readiness Engine (NVCRE) for certification, and NVSentinel for ongoing health monitoring.
AICR creates version‑locked recipes that can be deployed with Argo CD, Flux or Helm, while NVCRE runs NCCL tests and training workloads to verify GPU, network and storage readiness.
NVSentinel continuously watches GPU, node and network signals, can cordon or drain unhealthy nodes, and triggers revalidation with NVCRE after remediation.
All three components are released under the Apache 2.0 license and are intended to keep GPU clusters reliable throughout their lifecycle.
あなたにとっての意味
For AI practitioners, DSX OS offers a repeatable way to keep large GPU clusters healthy, reducing silent failures that waste compute. For users of free or locally hosted models, the platform could improve access to stable, certified infrastructure without relying on proprietary tooling.
ソース
- Configure, Certify, and Operate Kubernetes GPU Clusters with NVIDIA DSX OS, NVIDIA Developer (official), 5 Nov 2026
上記のソースから自動的に生成し、当社のコードでチェックしました：すべての数値と名前が含まれ、文がソースをコピーしてはいけません。間違いを見つけましたか？報告する。
今日無料で使えるAIはどれですか？
すべての無料AIモデルをブラインドヒューマン投票で順位付けし、あなたのGPUで必要なメモリと速度を示します。