--- title: NVIDIA open sources the full Run:ai platform for GPU... description: NVIDIA has open sourced the entire Run:ai platform for AI workload and GPU orchestration on Kubernetes, released as part of the NVIDIA DSX AI Factory Platform.... canonical: https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: NVIDIA open sources the full Run:ai platform for GPU orchestration on Kubernetes | daily.dev og:description: NVIDIA has open sourced the entire Run:ai platform for AI workload and GPU orchestration on Kubernetes, released as part of the NVIDIA DSX AI Factory Platform.... og:url: https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm og:image: https://api.daily.dev/og/posts/Av5zzKYQm.png og:image:alt: NVIDIA open sources the full Run:ai platform for GPU orchestration on Kubernetes og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # NVIDIA open sources the full Run:ai platform for GPU orchestration on Kubernetes **[Collections](https://daily.dev/sources/collections)** · 1 min read · 1 upvotes · 0 comments ## Summary NVIDIA has open sourced the entire Run:ai platform for AI workload and GPU orchestration on Kubernetes, released as part of the NVIDIA DSX AI Factory Platform. The platform splits into four independently adoptable components: KAI Scheduler (now a CNCF Sandbox project under Apache 2.0), GPU Fractioning Operator (CUDA 13.4-native GPU fractioning replacing the prior proprietary implementation), KAI Resource Management (projects, departments, quotas, node pools), and Workload Map (a Go library for managing Kubernetes workload types like PyTorchJob and RayCluster). Partners including Crusoe, Mirantis, Nebius, OpenNebula, Rafay, Saturn Cloud, Spectro Cloud, and vCluster are integrating the platform. Related NVIDIA open source projects Model Streamer and Grove were also mentioned. Commercial support remains available through NVIDIA Enterprise AI Factory offerings. ## Content NVIDIA has open sourced the entire Run:ai platform for AI workload and GPU orchestration on Kubernetes. It ships as part of the NVIDIA DSX AI Factory Platform. NVIDIA's AI infrastructure account announced it, and @ClementDelangue reposted the news. ## What's in the release The platform is split into four components, and each can be adopted on its own. - **KAI Scheduler** is now a CNCF Sandbox project under the Apache 2.0 license. - **GPU Fractioning Operator** brings CUDA 13.4-native GPU fractioning to Kubernetes. It replaces the previous proprietary implementation. - **KAI Resource Management** handles projects, departments, quotas, and node pools. - **Workload Map** is a Go library contract for managing different Kubernetes workload types, such as PyTorchJob and RayCluster. ## Partners Crusoe, Mirantis, Nebius, OpenNebula, Rafay, Saturn Cloud, Spectro Cloud, and vCluster are integrating the open platform into their own offerings. ## Related projects Two other NVIDIA open source projects were mentioned alongside the release. Model Streamer speeds up loading of model weights and is already integrated into vLLM, SGLang, and major clouds. Grove is an orchestration API for inference workloads made up of multiple components. ## Support Commercial support is still available through NVIDIA Enterprise AI Factory offerings. ## Questions this post answers ### What components make up NVIDIA's open sourced Run:ai platform? The platform consists of four independently adoptable components: KAI Scheduler, now a CNCF Sandbox project under Apache 2.0; GPU Fractioning Operator, which brings CUDA 13.4-native GPU fractioning to Kubernetes replacing the previous proprietary implementation; KAI Resource Management, handling projects, departments, quotas, and node pools; and Workload Map, a Go library contract for managing Kubernetes workload types like PyTorchJob and RayCluster. _Teams planning GPU scheduling on Kubernetes can follow rollout details like these on daily.dev._ ### Is NVIDIA's Run:ai GPU scheduler still proprietary software? No, the GPU Fractioning Operator and the rest of the Run:ai platform have been open sourced, replacing the previous proprietary CUDA GPU fractioning implementation with a CUDA 13.4-native version. The KAI Scheduler component specifically became a CNCF Sandbox project licensed under Apache 2.0, while commercial support remains available through NVIDIA Enterprise AI Factory offerings. _Developers evaluating licensing changes for GPU orchestration tools can track updates like this on daily.dev._ ### Which companies are integrating NVIDIA's open sourced Run:ai platform? Crusoe, Mirantis, Nebius, OpenNebula, Rafay, Saturn Cloud, Spectro Cloud, and vCluster are integrating the newly open Run:ai platform into their own cloud and infrastructure offerings. These integrations span GPU cloud providers and Kubernetes management platforms, signaling broad infrastructure ecosystem adoption of the orchestration stack. _Engineers comparing GPU cloud and Kubernetes vendors can follow integration news like this on daily.dev._ ## Similar posts on daily.dev - [The future of AI is community driven and open](https://daily.dev/posts/the-future-of-ai-is-community-driven-and-open-scdjbegms) · CNCF · 1 upvotes · 0 comments - [AI Infrastructure Runs On Kubernetes. Is Your Platform Ready?](https://daily.dev/posts/ai-infrastructure-runs-on-kubernetes-is-your-platform-ready--nnoyesuho) · Fairwinds Blog · 0 upvotes · 0 comments - [NVIDIA Is Putting Real Skin in the Open AI Game](https://daily.dev/posts/nvidia-is-putting-real-skin-in-the-open-ai-game-kzxum8vke) · Container Journal · 0 upvotes · 0 comments - [Streamline AI Infrastructure with NVIDIA Run:ai on Microsoft Azure](https://daily.dev/posts/streamline-ai-infrastructure-with-nvidia-run-ai-on-microsoft-azure-zjpzywais) · NVIDIA Developer · 0 upvotes · 0 comments --- Tags: [#open-source](https://daily.dev/tags/open-source), [#kubernetes](https://daily.dev/tags/kubernetes), [#nvidia](https://daily.dev/tags/nvidia) [View this post on daily.dev](https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"NVIDIA open sources the full Run:ai platform for GPU orchestration on Kubernetes","url":"https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm"},"datePublished":"2026-10-13T16:09:02.263Z","dateModified":"2026-10-13T16:09:36.585Z","description":"NVIDIA has open sourced the entire Run:ai platform for AI workload and GPU orchestration on Kubernetes, released as part of the NVIDIA DSX AI Factory Platform....","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"open-source,kubernetes,nvidia","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"NVIDIA open sources the full Run:ai platform for GPU orchestration on Kubernetes"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/nvidia-open-sources-the-full-run-ai-platform-for-gpu-orchestration-on-kubernetes-av5zzkyqm#faq","mainEntity":[{"@type":"Question","name":"What components make up NVIDIA's open sourced Run:ai platform?","acceptedAnswer":{"@type":"Answer","text":"The platform consists of four independently adoptable components: KAI Scheduler, now a CNCF Sandbox project under Apache 2.0; GPU Fractioning Operator, which brings CUDA 13.4-native GPU fractioning to Kubernetes replacing the previous proprietary implementation; KAI Resource Management, handling projects, departments, quotas, and node pools; and Workload Map, a Go library contract for managing Kubernetes workload types like PyTorchJob and RayCluster. Teams planning GPU scheduling on Kubernetes can follow rollout details like these on daily.dev."}},{"@type":"Question","name":"Is NVIDIA's Run:ai GPU scheduler still proprietary software?","acceptedAnswer":{"@type":"Answer","text":"No, the GPU Fractioning Operator and the rest of the Run:ai platform have been open sourced, replacing the previous proprietary CUDA GPU fractioning implementation with a CUDA 13.4-native version. The KAI Scheduler component specifically became a CNCF Sandbox project licensed under Apache 2.0, while commercial support remains available through NVIDIA Enterprise AI Factory offerings. Developers evaluating licensing changes for GPU orchestration tools can track updates like this on daily.dev."}},{"@type":"Question","name":"Which companies are integrating NVIDIA's open sourced Run:ai platform?","acceptedAnswer":{"@type":"Answer","text":"Crusoe, Mirantis, Nebius, OpenNebula, Rafay, Saturn Cloud, Spectro Cloud, and vCluster are integrating the newly open Run:ai platform into their own cloud and infrastructure offerings. These integrations span GPU cloud providers and Kubernetes management platforms, signaling broad infrastructure ecosystem adoption of the orchestration stack. Engineers comparing GPU cloud and Kubernetes vendors can follow integration news like this on daily.dev."}}]} ```