Microsoft Azure to Deploy AMD’s Helios AI Racks
Helios is AMD’s first integrated rack-scale AI system, combining Instinct GPUs, EPYC CPUs, Pensando networking and the ROCm software stack. Based on the Open Compute Project’s Open Rack Wide specification introduced by Meta, the platform shifts AMD’s competition with Nvidia beyond individual accelerators to complete data-center infrastructure, targeting a market where Nvidia has built a commanding position through its NVLink interconnects and CUDA software ecosystem.
AMD said on July 20, 2026, that Microsoft will deploy Helios on Azure to support frontier-model inference, customer workloads and Azure AI services, with shipments scheduled for the second half of 2026. Each rack integrates 72 Instinct MI455X GPUs and is designed to deliver as much as 2.9 exaFLOPS of FP4 performance and 31 terabytes of HBM4 memory. Financial terms of the expanded partnership were not disclosed.
All Coverage
2 original reportsThe Backstory
The history behind this eventAMD Unveils Helios Rack-Scale AI System With MI455X Accelerators to Challenge NVIDIA
AMD has long used its Instinct accelerators and ROCm software to challenge NVIDIA’s dominance in AI data centers. As large language model training and inference shift toward full-rack deployments, competition is expanding from individual chips to the integration of compute, memory and networking. Helios is therefore a key product in AMD’s push to win orders from cloud providers and large computing centers.
AMD unveiled Helios, its first rack-scale AI solution, at Computex 2026. Each rack contains 72 Instinct MI455X accelerators and uses HBM4 high-bandwidth memory for demanding LLM training and inference workloads. Its direct rival is NVIDIA’s next-generation Vera Rubin NVL72/VR200 platform, but AMD has yet to announce pricing or a formal shipping date.
Subscribe to Mark Radar Weekly
Every Friday, the week's strongest signals in your inbox. Unsubscribe anytime.