MATRIX · NVIDIA Run:ai https://derekwang2002.github.io/ai-intelligence/en/projects/nvidia-runai/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT NVIDIA documents Workload Map architecture and three reuse paths https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261013-02/ ev-20261013-02-workload-map Mon, 02 Nov 2026 16:56:00 GMT Update TRIAL <p>NVIDIA documents Workload Map architecture and three reuse paths</p><p>Original story: NVIDIA extends Run:ai with GPU controls and Workload Map reuse</p><p><strong>Why read: </strong>A common workload map can reduce platform adapters, but adoption and interoperability remain mainly vendor-supported.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://developer.nvidia.com/blog/support-new-ai-frameworks-faster-using-nvidia-workload-map/">Support New AI Frameworks Faster Using NVIDIA Workload Map</a> · <a href="https://github.com/dsx-ai-factory/workload-map">DSX AI Factory Workload Map repository</a> · <a href="https://github.com/dsx-ai-factory/workload-map/releases/tag/v0.3.0">Workload Map v0.3.0 release</a></p> Run:ai adds CUDA 13.4 memory-enforced GPU fractioning https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261013-02/ ev-20261013-02-gpu-fractioning Wed, 28 Oct 2026 15:58:44 GMT Update TRIAL <p>Run:ai adds CUDA 13.4 memory-enforced GPU fractioning</p><p>Original story: NVIDIA extends Run:ai with GPU controls and Workload Map reuse</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://developer.nvidia.com/blog/memory-enforced-gpu-sharing-in-kubernetes-with-nvidia-cuda-13-4/">Memory-Enforced GPU Sharing in Kubernetes with NVIDIA CUDA 13.4</a> · <a href="https://github.com/kai-scheduler/gpu-fractioning">KAI GPU Fractioning Operator</a></p> NVIDIA extends Run:ai with GPU controls and Workload Map reuse https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261013-02/ ev-20261013-02-published Wed, 14 Oct 2026 15:57:53 GMT New TRIAL <p>NVIDIA has combined memory-enforced GPU fractioning in Run:ai with a reusable Workload Map layer for heterogeneous Kubernetes AI workloads. Workload Map converts supported CRDs into a common tree used by Run:ai, KAI Scheduler and, according to NVIDIA, Elotl Nova. The library is Apache-2.0 and released, but production-use and interoperability evidence remains vendor-reported.</p><p><strong>Why read: </strong>The complete Run:ai GPU orchestration stack is now Apache-2.0 and inspectable, while the community build carries no enterprise support.</p><p>Recommendation: TRIAL</p><p>Sources: <a href="https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform/">NVIDIA Run:ai is now open source</a> · <a href="https://github.com/dsx-ai-factory/run-ai">Run:ai open-source platform repository</a></p>