RELEASE · COMPANIES · #1907
NVIDIA open-sources Run:ai platform as part of DSX AI Factory
NVIDIA has open-sourced the full Run:ai platform as part of its DSX AI Factory Platform, publishing the code on GitHub and offering it as a Community Product on NGC. The release includes four modular components (KAI Scheduler, GPU Fractioning Operator, KAI Resource Management, Workload Map), complements projects like Model Streamer and Grove, and is being integrated by multiple DSX ecosystem partners such as Crusoe, Mirantis and Nebius.
KEY POINTS
- NVIDIA has open-sourced the full Run:ai platform as part of its DSX AI Factory Platform, publishing the code on GitHub and offering it as a Community Product on NGC.
- The release includes four modular components (KAI Scheduler, GPU Fractioning Operator, KAI Resource Management, Workload Map), complements projects like Model Streamer and Grove, and is being integrated by multiple DSX ecosystem partners such as Crusoe, Mirantis and Nebius.
- Open-sourcing a production-grade GPU and AI workload orchestration platform lets partners and clouds adopt, extend, and integrate NVIDIA’s orchestration innovations faster, potentially improving GPU utilization and accelerating AI deployments.
WHY IT MATTERS
Open-sourcing a production-grade GPU and AI workload orchestration platform lets partners and clouds adopt, extend, and integrate NVIDIA’s orchestration innovations faster, potentially improving GPU utilization and accelerating AI deployments.