{"slug": "amd-expands-into-rack-scale-ai-as-cpu-demand-rises", "title": "AMD expands into rack-scale AI as CPU demand rises", "summary": "AMD unveiled its Helios rack-scale AI system at its Advancing AI 2026 conference on July 23, a fully integrated rack packing 72 Instinct MI455X GPUs and 18 6th Gen EPYC \"Venice\" CPUs with up to 2.9 exaflops of peak FP4 performance and 31 TB of HBM4 memory. AMD claims Helios delivers up to 30% more tokens per dollar than Nvidia's Rubin NVL72, with Anthropic committing up to 2 GW of MI450-series GPUs, OpenAI and Meta each signing 6 GW deals, and shipments expected to begin in the second half of 2026 and ramp through 2027. AMD also projected the server CPU market will surpass $200 billion by 2030 and reported a fourfold increase in AI energy efficiency from 2024 through mid-2026 toward its 20x-by-2030 goal.", "body_md": "Logo via Wikimedia Commons; treatment-A cover, license to verify on approval\n\n# AMD expands into rack-scale AI as CPU demand rises\n\nThe chipmaker unveiled its Helios system at Advancing AI 2026, taking aim at Nvidia's dominance with a full-rack solution designed for agentic workloads.\n\n[AMD](https://cryptobriefing.com/markets/amd/) is no longer content selling chips. It wants to sell the whole rack.\n\nAt its Advancing AI 2026 conference on July 23, the company revealed the Helios rack-scale AI system, a fully integrated solution that bundles GPUs, CPUs, networking, and software into a single deployable unit.\n\n## What Helios actually is\n\nEach Helios rack packs 72 Instinct MI455X GPUs alongside 18 6th Gen EPYC “Venice” server CPUs, connected through AMD’s Pensando networking hardware and managed by its ROCm software stack. The result is a system capable of up to 2.9 exaflops of peak FP4 performance, with 31 TB of HBM4 memory.\n\nAMD claims the Helios system delivers up to 30% more tokens per dollar than [Nvidia](https://cryptobriefing.com/markets/nvidia/)’s Rubin NVL72 rack.\n\nThe system is designed for two primary use cases: frontier model training and large-scale inference. Agentic AI workloads, where models autonomously execute multi-step reasoning and orchestration tasks, place unusually heavy demands on CPUs rather than just GPUs.\n\n## The customer list tells the story\n\nAnthropic has committed to deploying up to 2 GW worth of MI450-series GPUs in Helios racks. OpenAI has signed on for a 6 GW multigenerational deal. [Meta](https://cryptobriefing.com/markets/meta/) has matched that with its own 6 GW commitment. [Microsoft](https://cryptobriefing.com/markets/microsoft/) and [Oracle](https://cryptobriefing.com/markets/oracle/) round out the roster.\n\nShipments of the Helios system are expected to begin in the second half of 2026 and ramp through 2027.\n\n## The server CPU market gets a new narrative\n\nAMD is projecting that the server CPU market will surpass $200 billion by 2030. [Bank of America](https://cryptobriefing.com/markets/bank-of-america/) has its own forecast exceeding $170 billion, driven largely by the same agentic AI dynamics AMD is targeting.\n\nAMD reported achieving a fourfold increase in AI energy efficiency from 2024 through mid-2026, well ahead of schedule for its stated goal of reaching a 20x improvement by 2030.\n\nThe company also open-sourced its Optima solution for optimizing internal AI processes.\n\n**Disclosure:** This article was edited by Editorial Team. For more information on how we create and review content, see our\n\n[Editorial Policy](https://cryptobriefing.com/editorial-policy/).", "url": "https://wpnews.pro/news/amd-expands-into-rack-scale-ai-as-cpu-demand-rises", "canonical_source": "https://cryptobriefing.com/amd-helios-rack-scale-ai-system/", "published_at": "2026-09-11 17:15:26+00:00", "updated_at": "2026-09-11 17:44:16.438926+00:00", "lang": "en", "topics": ["ai-infrastructure", "ai-chips", "ai-products", "ai-agents"], "entities": ["AMD", "Helios", "Instinct MI455X", "EPYC Venice", "Nvidia", "Anthropic", "OpenAI", "Meta"], "alternates": {"html": "https://wpnews.pro/news/amd-expands-into-rack-scale-ai-as-cpu-demand-rises", "markdown": "https://wpnews.pro/news/amd-expands-into-rack-scale-ai-as-cpu-demand-rises.md", "text": "https://wpnews.pro/news/amd-expands-into-rack-scale-ai-as-cpu-demand-rises.txt", "jsonld": "https://wpnews.pro/news/amd-expands-into-rack-scale-ai-as-cpu-demand-rises.jsonld"}}