Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
[forminator_form id="25163"]

blogs.nvidia.com.techcommunity.microsoft.com+1.blogs.nvidia.com.NVIDIA and Microsoft announced a wider partnership at Microsoft Ignite on Thursday, November 19, 2026. The deal covers Azure data centers, on-premises and edge systems, and Windows PCs. Azure is expanding access to virtual machines built on NVIDIA Vera Rubin NVL72 systems, Microsoft Foundry is adding NVIDIA open models and customization tools, and Azure Local is gaining support for more NVIDIA hardware. Together, the moves give businesses more ways to build and run AI in the cloud or close to where their data lives.blogs.nvidia.com
According to NVIDIA, Vera Rubin NVL72 systems are now running in Azure data centers to handle reasoning, agentic workloads, and large-scale training and inference. The rollout has been under way for most of the year. Microsoft says it was the first cloud to bring up a Vera Rubin NVL72 system for validation, in March, and that it delivered the first production systems to its data centers in August.techcommunity.microsoft.com+1
The two companies are also working to extend confidential computing to bare-metal and rack-scale AI systems. At Ignite, Microsoft showed a confidential bare-metal demo that ran a 397-billion-parameter model on one node with four Rubin GPUs. Confidential virtual machines and Foundry integration for these systems are still in development.techcommunity.microsoft.com
Foundry Managed Compute now supports an expanded NVIDIA Nemotron 3.5 lineup. Nemotron 3.5 Super is designed for long-running agents, while Lightning targets fast responses and tool use. Bolt and Transcribe models are coming soon. NVIDIA said EY used Microsoft's serverless Programmed Training service to adapt Nemotron models for classifying corporate purchasing records. EY reported a 55% gain in accuracy and costs 15 times lower than proprietary frontier models. Microsoft's Ignite Book of News also says customers will be able to deploy open-weight models and agents from Foundry to their own Azure Kubernetes Service clusters.news.microsoft.com+1
Azure Local, Microsoft's on-premises platform, will support NVIDIA RTX PRO 6000 Blackwell GPUs from hardware partners. It is also expanding to NVIDIA's Jetson Thor and Jetson Orin edge platforms for uses like machine vision and robotics. In addition, Foundry Local is getting Nemotron 3.5 and NVIDIA's Cosmos physical AI models.blogs.nvidia.com
On the PC side, the companies are demonstrating local large language models running on Windows PCs with NVIDIA RTX Spark. They are also adding Direct Stream to GPU-accelerated Windows 365 Cloud PCs, which supports up to two 4K streams at 60 frames per second. A limited preview is open now, and a public preview is planned for early 2027.blogs.nvidia.com
Microsoft is also using NVIDIA technology for its own products. Microsoft AI uses NVIDIA Dynamo for inference during reinforcement learning post-training, and Bing runs its search models on RTX PRO 6000 Blackwell GPUs. Ignite runs through Friday, November 20, at San Francisco's Moscone Center and online.techcommunity.microsoft.com+1