NVIDIA open-sources its full Run:ai GPU orchestration platform
NVIDIA released the complete Run:ai platform as open-source software as part of its DSX AI Factory platform, along with four components that can be adopted separately. Eight cloud and infrastructure partners, including Crusoe and Nebius, are integrating it.
Photo: Coolcaesar / Wikimedia Commons, CC BY-SA 4.0. NVIDIA logo via Wikimedia Commons (Apache License 2.0).
NVIDIA said on October 13 that it is making the full NVIDIA Run:ai platform available as open-source software. Run:ai, which NVIDIA says helps organizations run AI workloads on Kubernetes by dynamically allocating GPUs and managing jobs from data processing and training to inference, becomes an open foundation of the company's DSX AI Factory platform.
The release includes the integrated platform plus four components that can be used on their own: KAI Scheduler, the Kubernetes-native scheduling engine NVIDIA had already released under the Apache 2.0 license and contributed to the CNCF as a Sandbox project; a GPU Fractioning Operator that lets workloads share GPUs using the fractioning built into CUDA 13.4; KAI Resource Management, for team quotas, organizational hierarchies and node pools; and Workload Map, a Go library that gives tools a consistent way to describe different kinds of Kubernetes AI workloads.
NVIDIA says the source code is in its DSX AI Factory GitHub organization and that the platform is also on NVIDIA NGC as a Community Product with a Helm chart. No paid Run:ai subscription is required for the Community Product, but NVIDIA enterprise support is not included. Eight partners, Crusoe, Mirantis, Nebius, OpenNebula Systems, Rafay, Saturn Cloud, Spectro Cloud and vCluster, are integrating the open-source platform into their offerings.
Run:ai remains available through NVIDIA-supported commercial offerings, including as a standalone product and through NVIDIA Mission Control and NVIDIA AI Enterprise, the company says.