--- type: "post" id: "nvidia-s-gpu-sharing-software-is-now-free-8c9a45" slug: "nvidia-s-gpu-sharing-software-is-now-free-8c9a45" title: "Nvidia's GPU-sharing software is now free and open source : the full Run:ai platform, which helps companies split GPUs between AI jobs, is now on GitHub for anyone to use." description: "Thesis: Open-sourcing Run:ai moves GPU workload orchestration from a primarily vendor-delivered platform toward a set of inspectable and independently adoptable components." retrieval_nugget: "Thesis: Open-sourcing Run:ai moves GPU workload orchestration from a primarily vendor-delivered platform toward a set of inspectable and independently adoptable components." published_at: "2026-10-14" updated_at: "2026-10-14" record_date: "2026-10-14" date_kind: "published_at" topics: [] source_urls: ["https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform"] schema_version: "newruntime-agent-readable-v0.2" stable_id: "post:nvidia-s-gpu-sharing-software-is-now-free-8c9a45" status: "published" visuals: [] editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:421188d3506a0121196354687db70725867709f8dcbe1b4c04f369880b58926b","reviewed_at":"2026-10-14T11:02:50.813Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":1,"related_record_count":0} analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Open-sourcing Run:ai moves GPU workload orchestration from a primarily vendor-delivered platform toward a set of inspectable and independently adoptable components.","observed_facts":[{"text":"NVIDIA says it is releasing the full Run:ai platform as open-source software as part of its DSX AI Factory platform.","source_urls":["https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform"]},{"text":"The release lists four independently adoptable components: KAI Scheduler, GPU Fractioning Operator, KAI Resource Management, and Workload Map.","source_urls":["https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform"]},{"text":"NVIDIA says NGC Community Product access does not require a paid Run:ai subscription or support entitlement, but enterprise support is not included.","source_urls":["https://developer.nvidia.com/blog/nvidia-runai-is-now-open-source-as-part-of-nvidia-dsx-ai-factory-platform"]}],"mechanism":"Modular availability allows teams to adopt the integrated platform or evaluate and integrate individual orchestration components into an existing infrastructure stack.","why_now":"NVIDIA presents the release as an open foundation for its DSX ecosystem and says partners can build AI factory offerings on it.","implications":["Infrastructure teams can inspect and test individual components without committing to the whole Run:ai platform.","The open-source release does not by itself establish support parity, production adoption, or measurable GPU-utilization gains."],"evidence_boundary":"The source is NVIDIA's announcement; it describes availability and terms but does not independently verify deployments, performance, or adoption outcomes.","watch_conditions":["Look for deployments or partner offerings that document which components they use and what support and licensing terms apply."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this release opens a distinct analytical record on access and modularity in GPU workload orchestration."} editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]} routes: {"html":"https://newruntime.com/posts/nvidia-s-gpu-sharing-software-is-now-free-8c9a45/","markdown":"https://newruntime.com/posts/nvidia-s-gpu-sharing-software-is-now-free-8c9a45.md","json":"https://newruntime.com/posts/nvidia-s-gpu-sharing-software-is-now-free-8c9a45.json"} --- # Nvidia's GPU-sharing software is now free and open source : the full Run:ai platform, which helps companies split GPUs between AI jobs, is now on GitHub for anyone to use. ## Retrieval answer Thesis: Open-sourcing Run:ai moves GPU workload orchestration from a primarily vendor-delivered platform toward a set of inspectable and independently adoptable components. Thesis: Open-sourcing Run:ai moves GPU workload orchestration from a primarily vendor-delivered platform toward a set of inspectable and independently adoptable components. NVIDIA states that it is releasing the full Run:ai platform as open-source software as part of its DSX AI Factory platform. The release includes four components that can be adopted independently: KAI Scheduler, GPU Fractioning Operator, KAI Resource Management, and Workload Map. NVIDIA also says the platform can be deployed directly from its GitHub organization or through an NGC Community Product. These facts matter because the change is not merely a repository publication of one utility; it makes a broader orchestration stack available for inspection and integration. Mechanism: Modular availability allows teams to adopt the integrated platform or evaluate and integrate individual orchestration components into an existing infrastructure stack. The practical implication is that teams with existing Kubernetes and GPU operations can evaluate or compose parts without first committing to the whole Run:ai product. Why now: NVIDIA presents the release as an open foundation for its DSX ecosystem and says partners can build AI factory offerings on it. That is the company’s stated rationale, not independent evidence of adoption or improved utilization. Evidence boundary: The source is NVIDIA's announcement; it describes availability and terms but does not independently verify deployments, performance, or adoption outcomes. NVIDIA says a paid subscription or support entitlement is not required to use the NGC Community Product, but enterprise support is not included, and applicable open-source licenses and product terms vary by offering. The announcement alone does not establish deployment scale, performance gains, or the degree of operational independence from NVIDIA software and hardware. First implication: Infrastructure teams can inspect and test individual components without committing to the whole Run:ai platform. First watch condition: Look for deployments or partner offerings that document which components they use and what support and licensing terms apply. Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json ## Continue in the graph Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`. - collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis. - collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph. - whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)