NVIDIA said it is releasing the full NVIDIA Run:ai platform as open source software as part of its DSX AI Factory Platform, giving customers and partners a foundation for AI workload and GPU orchestration. The release covers the integrated platform plus four components that can be adopted separately: KAI Scheduler for GPU-aware Kubernetes scheduling, the GPU Fractioning Operator for CUDA-native GPU sharing, KAI Resource Management for organizational quota and hierarchy, and Workload Map for a consistent workload-management contract. NVIDIA said eight DSX ecosystem partners are integrating the open source platform into their offerings, including Crusoe with Managed Kubernetes, Mirantis extending k0rdent AI, Nebius offering one-click deployment in its AI Cloud, OpenNebula Systems wiring it into its AI Cloud Platform through OneKS, and Rafay, Saturn Cloud, Spectro Cloud and vCluster. The complete source is published in the NVIDIA Run:ai repository in the NVIDIA DSX AI Factory GitHub organization and on NGC as a Community Product with a Helm chart, and NVIDIA said no paid Run:ai subscription or NVIDIA support entitlement is required for it, though enterprise support is not included and the NGC download requires accepting NVIDIA’s software license terms. CTech noted the move comes about two years after NVIDIA’s roughly $800 million acquisition of the Israeli-founded company, and NVIDIA pointed to the related open source projects Model Streamer and Grove.
NVIDIA releases the full Run:ai platform as open source
NVIDIA said it is releasing the full NVIDIA Run:ai platform as open source software as part of its DSX AI Factory Platform, giving customers and partners a foundation for AI workload and GPU orchestration. The release covers the integrated platform plus four components that can be adopted separately: KAI Scheduler for GPU-aware Kubernetes scheduling, the GPU Fractioning Operator for CUDA-native GPU sharing, KAI Resource Management for organizational quota and hierarchy, and Workload Map for a consistent workload-management contract. NVIDIA said eight DSX ecosystem partners are integrating the open source platform into their offerings, including Crusoe with Managed Kubernetes, Mirantis extending k0rdent AI, Nebius offering one-click deployment in its AI Cloud, OpenNebula Systems wiring it into its AI Cloud Platform through OneKS, and Rafay, Saturn Cloud, Spectro Cloud and vCluster. The complete source is published in the NVIDIA Run:ai repository in the NVIDIA DSX AI Factory GitHub organization and on NGC as a Community Product with a Helm chart, and NVIDIA said no paid Run:ai subscription or NVIDIA support entitlement is required for it, though enterprise support is not included and the NGC download requires accepting NVIDIA's software license terms. CTech noted the move comes about two years after NVIDIA's roughly $800 million acquisition of the Israeli-founded company, and NVIDIA pointed to the related open source projects Model Streamer and Grove.
Source: Nvidia