Lambda to make open-source Run:ai scheduler a managed option on its Kubernetes service

Lambda said on October 13, 2026 that it is working with Nvidia to make the open-sourced Run:ai GPU scheduler a first-class managed option on Lambda Kubernetes Service (LKS), part of a planned library of managed schedulers. Lambda has supported Run:ai since 2022 and plans to add Slurm and other schedulers validated against its own hardware.

• Run:ai will ship as a managed component on LKS alongside other schedulers
• Nvidia open-sourced the KAI Scheduler under Apache 2.0 in April 2025
• The scheduler offers GPU sharing and quota-based scheduling with borrowing
• LKS already auto-remediates nodes on XID errors and NVLink degradation