NVIDIA has released the full NVIDIA Run:ai platform as open-source software under the Apache 2.0 license on GitHub. This release places production GPU scheduling, quotas, and workload management directly into the hands of developers.
Key points
- NVIDIA Run:ai platform is now open source under Apache 2.0.
- Developers gain control over GPU scheduling and workload management.
- Platform integrates into NVIDIA DSX AI Factory.
- Run:ai orchestrates AI workloads on Kubernetes clusters.
The open-sourced Run:ai platform is now part of the NVIDIA DSX AI Factory platform. This allows platform teams to inspect, fork, and self-host the stack without requiring a paid subscription.
NVIDIA Run:ai is an AI workload and GPU orchestration platform designed for Kubernetes. It schedules development, training, and inference workloads across shared GPU clusters, running on top of Kubernetes without replacing it. The control plane manages workload submission, resource management, and monitoring, while cluster services add GPU-aware scheduling.
The platform is accessible via a web UI, CLI, REST API, and an MCP server for AI assistants. It is composed of four modular components that teams can adopt independently: the KAI Scheduler, a Kubernetes-native scheduling engine and CNCF Sandbox project; and the GPU Fractioning Operator, which brings CUDA 13.4-native GPU fractions to Kubernetes.
Sources and references
Original report
NVIDIA Open Sources Run:ai Under Apache 2.0 for Kubernetes GPU Orchestration
MarkTechPost
Read on MarkTechPost (opens in a new tab)Linked in the original report
- Codegithub.com/dsx-ai-factory/run-ai (opens in a new tab)
- Codegithub.com/kai-scheduler/KAI-Scheduler (opens in a new tab)
- Codegithub.com/cncf/sandbox/issues/372 (opens in a new tab)
- Codegithub.com/kai-scheduler/kai-gpu-fractioning (opens in a new tab)
- Codegithub.com/kai-scheduler/kai-resource-management (opens in a new tab)
- Codegithub.com/dsx-ai-factory/karta (opens in a new tab)
Ref 00OZT619
Cite this
APA
Interestana AI Editorial. (2026, October 19). NVIDIA Open Sources Run:ai for Kubernetes GPU Orchestration. Interestana. https://interestana.com/articles/nvidia-open-sources-runai-under-apache-20-for-kubernetes-gpu-orchestration-00ozt619
MLA
“NVIDIA Open Sources Run:ai for Kubernetes GPU Orchestration.” Interestana, 19 Oct. 2026, interestana.com/articles/nvidia-open-sources-runai-under-apache-20-for-kubernetes-gpu-orchestration-00ozt619.
Chicago
Interestana AI Editorial. “NVIDIA Open Sources Run:ai for Kubernetes GPU Orchestration.” Interestana, October 19, 2026. https://interestana.com/articles/nvidia-open-sources-runai-under-apache-20-for-kubernetes-gpu-orchestration-00ozt619.
BibTeX
@misc{interestana_00ozt619,
author = {{Interestana AI Editorial}},
title = {{NVIDIA Open Sources Run:ai for Kubernetes GPU Orchestration}},
howpublished = {Interestana},
year = {2026},
month = oct,
url = {https://interestana.com/articles/nvidia-open-sources-runai-under-apache-20-for-kubernetes-gpu-orchestration-00ozt619},
note = {Summary of reporting by MarkTechPost, https://www.marktechpost.com/2026/10/18/nvidia-runai-open-source-gpu-orchestration-kubernetes/}
}Credit the source
Original reporting: MarkTechPost, “NVIDIA Open Sources Run:ai Under Apache 2.0 for Kubernetes GPU Orchestration”, https://www.marktechpost.com/2026/10/18/nvidia-runai-open-source-gpu-orchestration-kubernetes/(opens in a new tab)
Follow this story
Get Computing & Chips in your inbox — new stories in 40 words or fewer, once a week.
Subscribe