01 What happened
NVIDIA has released the complete Run:ai source code as part of DSX AI Factory. The platform orchestrates AI workloads and GPU resources on Kubernetes. Teams can adopt the integrated stack or use its individual components.
02 Key details
- The four independently adoptable components are KAI Scheduler, GPU Fractioning Operator, KAI Resource Management and Workload Map.
- NVIDIA explicitly identifies Apache 2.0 as the license for KAI Scheduler. The rest of the source code is governed by its applicable open-source licenses.
- GPU Fractioning Operator exposes CUDA 13.4-native GPU fractioning in Kubernetes. The driver limits memory per workload, replacing the previous proprietary implementation.
- The NGC Community Product includes a Helm chart. No paid Run:ai subscription or support entitlement is required, but NVIDIA license and product-term acceptance is mandatory; enterprise support is not included.
03 Why it matters
AI infrastructure teams can inspect and modify the code, allocate capacity across departments and adopt only the components they need. Teams choosing the NGC installation should review its license terms and arrange support separately: the Community Product does not include NVIDIA enterprise support.
04 Who it matters to
AI infrastructure engineers, cluster administrators and machine learning teams.
Original source