Nvidia open-sources Run:ai GPU orchestration platform for Kubernetes

On October 13, 2026, Nvidia released the full Run:ai platform as open source under its new DSX AI Factory GitHub organization. The release includes the KAI Scheduler, a GPU Fractioning Operator, KAI Resource Management and the Workload Map library, with eight partners announcing integrations.

• KAI Scheduler is Apache 2.0 with gang scheduling, quotas and GPU sharing via MIG, MPS and time slicing
• GPU Fractioning Operator wires CUDA-native GPU fractions into Kubernetes and is scheduler-agnostic
• Nvidia retired its proprietary fractional-GPU code in favor of native fractions in CUDA 13.4
• A Helm-installable Community Product build is on NGC, but enterprise support is not included