AMD and Microsoft have expanded their long-term partnership to deploy AMD’s next-generation AI infrastructure across Microsoft Azure, with the companies collaborating on processors, accelerators, networking and software to support large-scale artificial intelligence workloads.
As part of the agreement, Microsoft will deploy AMD’s Helios Rackscale Solution, which combines AMD Instinct MI455X GPUs, sixth-generation AMD EPYC “Venice” processors, Pensando networking technology and ROCm software. According to AMD, the platform will be used to support AI inference for Microsoft’s frontier AI models, Azure AI services and enterprise customers running AI workloads on Azure.
AMD said customer shipments of the Helios platform, including deployments for Microsoft, are expected to begin in the second half of 2026.
In addition to the Helios deployment, Microsoft will introduce two new Azure virtual machine (VM) series powered by sixth-generation AMD EPYC processors. The Azure HDv2 series is designed for agentic AI and data processing workloads, while the Azure HXv2 series targets semiconductor design and engineering applications.
The partnership also extends to Azure’s networking infrastructure. Microsoft will expand its deployment of AMD Pensando data processing units (DPUs) and integrate AMD technologies with Azure Boost to improve networking efficiency and data processing across its cloud platform.
Commenting on the announcement, AMD Chair and Chief Executive Officer Lisa Su said the expanded collaboration builds on the companies’ long-standing work in cloud infrastructure and broadens the deployment of AMD’s AI technologies across Azure. Microsoft Chairman and Chief Executive Officer Satya Nadella said customers increasingly require AI infrastructure capable of supporting a wide range of workloads, from model training and inference to data preparation and reinforcement learning.
The announcement comes as cloud providers continue to expand AI infrastructure investments to meet growing enterprise demand for generative AI services, large language models and AI-driven applications.