AMD Helios launches with Microsoft, shakes up AI market

AMD’s Helios AI platform ships in H2 2026, with Microsoft deploying it on Azure. The partnership boosts AMD’s stock and positions it as a major Nvidia rival in AI infrastructure.

AMD will begin shipping its Helios rack-scale AI platform in the second half of 2026, with Microsoft as the first major customer deploying it at scale on Azure for AI inference. Helios features AMD’s next-generation Instinct MI455X GPUs, 6th-generation EPYC Venice CPUs, Pensando networking, and the ROCm software stack, delivering up to 1.4 exaFLOPS of FP8 compute per rack. Positioned as a direct competitor to Nvidia’s rack-scale solutions, each Helios rack is priced between $5 million and $5.5 million. Microsoft’s adoption of Helios has driven a surge in AMD’s stock price, with analysts estimating AMD could capture up to 25% of the GPU market. Azure will also launch new AMD-based virtual machines for AI inference and semiconductor design, expanding its hardware options. This collaboration marks a significant shift in the AI infrastructure landscape. Meta is also planning to use custom MI450-based GPUs for AI workloads starting in late 2026. While specific deployment volumes remain undisclosed, engineering samples and low-volume production will start in H2 2026, with mass production expected by Q2 2027.