TensorWave Powers Frontier AI Growth for Cloud Customers with AMD Helios Rackscale Solution
TensorWave adopts AMD Helios rackscale solution with 72 MI455X GPUs delivering 2.9 exaFLOPS peak and 31 TB HBM4 per rack for large-scale training and inference. The open rackscale architecture with integrated networking (Pensando Vulcano) and ROCm software enables better cost-per-token economics and clustering efficiency for frontier model workloads.
Microsoft will use AMD’s AI-optimized Helios racks in Azure - SiliconANGLE
Company has already tested the beefy AI rack
AMD launched Helios, a rack-scale system for frontier models, with OpenAI, Meta, Microsoft, and Anthropic for gigawatt-scale deployment. The system breaks Nvidia dominance, forcing GPU diversification as agentic AI drives the market to $1.4 trillion by 2030.
AMD launches Helios, integrated rack-scale system with 72 MI455X GPUs ($5–5.5M per rack); Q1 revenue $5.8B (+57% YoY). Anthropic deal and 30% cost-per-inference edge validate credible Nvidia alternative, reshaping enterprise GPU procurement.
AMD repositions as systems company with Helios, integrating GPU/CPU/networking into rack-scale infrastructure; ZT Systems acquisition accelerates delivery. GPU market shifting from components to integrated systems; memory bandwidth critical differentiator for consolidating AI workloads onto fewer GPUs.
AMD lancia Helios: 72-GPU MI455X con 30% price/performance lead su Nvidia Vera Rubin. Per la prima volta sincronizzato col competitor dopo anni di gap: ristabilisce concorrenza nel datacenter AI, riaprendo scelte budget e architetture ai buyer enterprise.
AMD unveils Helios platform: 72-GPU rack with MI455X (432GB HBM4, 2.9 exaflops) and Venice CPUs for reasoning models and agentic workflows. Signals viable NVIDIA challenger for hyperscalers; open UALink networking vs Vera Rubin's proprietary stack.
Following ZT acquisition, company looks to compete with Nvidia's system-level approach
AMD is making a serious play for Nvidia's crown with its new Helios rack-scale AI platform, Venice CPUs, and MI455X chips.
AMD challenges Nvidia’s AI dominance with Helios, new Instinct GPUs, EPYC processors, ROCm.ai software, and major customer deployment plans.
Spec for spec, the House of Zen's first rack-scale AI compute platform is bigger and faster than Nvidia's Vera Rubin by nearly every metric, but that's only on paper
AMD launched the MI455X accelerator, Helios server rack, and Venice Epyc processor at its Advancing AI event, predicting a $2 trillion market by 2030
Massive performance gains, higher memory capacity, and a large scale-up domain make for AMD’s strongest AI chip yet
Vultr Scales Next-Generation AI Infrastructure with AMD Helios Rackscale Solution Powered by AMD Instinct™ MI455X GPUs
AMD is challenging its chipmaker rival with a new rack-scale system that will start shipping to customers later this year.
AMI Announces AMI Rack Manager Support for AMD Helios Rackscale Solution and Aptio V and MegaRAC OneTree Support for 6th Gen AMD EPYC Processors, Accelerating Deployment for AI…
AMD e Cerebras hanno annunciato una partnership che le vedr� combinare la piattaforma rack-scale Helios con il Wafer-Scale Engine (WSE) per realizzare un'infrastruttura di…
AMD introduces its Helios rack-scale AI solution and sixth-gen Epyc CPUs, targeting Nvidia's lead in AI infrastructure and promising shipments by Q3 2026.
Next-generation data center GPUs and modular rack architectures power frontier AI training and their inference workloads.