Nvidia Run Is Now Open Source: What’s Free and Paid

Nvidia has recently made its entire Run:ai GPU orchestration platform open source. This means that IT teams can now manage GPU scheduling and AI workloads on Kubernetes without having to pay for a software subscription. The open-source version of Run:ai provides essential features such as GPU scheduling, resource management, and AI workload orchestration for Kubernetes.

To deploy the software in a production environment, teams can utilize the self-hosted control plane that comes with the open-source version, rather than purchasing a Run subscription. However, they will be responsible for managing their own infrastructure, upgrades, and any operational issues that arise.

The open-source release of Run:ai is available under the Apache 2.0 license, allowing organizations to inspect, modify, and build upon the software. There are four modular components available alongside the main platform: KAI Scheduler, GPU Fractioning Operator, and integrated platform through Nvidia NGC open-source collection.

While enterprise support remains a paid option, organizations using the free version must manage all aspects of their infrastructure, upgrades, and operational tasks. This distinction is crucial for companies investing in larger AI hardware deployments as they still need to provide Kubernetes clusters, compatible GPUs, storage, networking, monitoring, and administrators to maintain the platform.

Nvidia has not provided a universal price for the supported Run configurations, as the cost of operation depends on factors such as existing infrastructure, staffing, and support requirements. Organizations should review Nvidia's documentation to understand the deployment options and the distinction between supported and unsupported installations.