AMD Unveils Helios, Its Next-Gen AI Powerhouse With MI455X & 6th Gen EPYC, Challenging NVIDIA's Rack-Scale Dominance

Key Points:

• AMD has unveiled its next-generation Helios AI Rack, a fully liquid-cooled, open-standards platform combining AMD Instinct MI455X GPUs, 6th Gen EPYC Venice CPUs, and Pensando AI networking chips to target large-scale AI training and inference workloads.
• The Helios AI Rack features 72 MI455X GPUs delivering up to 2.9 Exaflops FP4 compute and 31 TB of HBM4 memory, alongside 256-core Zen 6 EPYC CPUs fabricated on TSMC’s 2nm process, promising significant performance and efficiency improvements over competitors.
• Networking is enabled by AMD Pensando "Vulcano" 800 AI NICs and "Salina" DPUs, providing high bandwidth, low latency, and advanced offloading capabilities to support scalable AI cluster interconnects with up to 43 TB/s scale-out bandwidth.
• AMD’s ROCm software stack supports major AI frameworks with Day-0 compatibility, and early customers include Microsoft (for Azure AI services), OpenAI, Meta, Oracle, and others, signaling strong industry adoption ahead of the official launch at Advancing AI 2026.
• Priced around $5–5.5 million per rack and consuming approximately 225-245kW, Helios aims to disrupt NVIDIA’s market dominance by offering higher memory density, superior performance-per-dollar, and an open, interoperable AI infrastructure for trillion-parameter model training and inference.