[AMD Launches First Rack-Scale AI System Helios, Microsoft Joins Procurement Alliance to Challenge NVIDIA's AI Compute Dominance] BlockBeats News, July 20 — AMD officially launched its first rack-scale system for artificial intelligence, Helios, and announced that Microsoft will deploy the system in its Azure data centers, becoming another major customer after Meta, OpenAI, Oracle, and others. Helios is regarded as a key product in AMD's recent efforts to challenge NVIDIA's dominance in AI infrastructure and is the first AI rack solution to directly compete with NVIDIA's Grace Hopper and Vera Rubin systems. It is expected to begin delivery later this year. Microsoft stated that it will leverage AMD Helios to support cutting-edge model inference, AI application development, and Azure AI services, while simultaneously launching two types of compute instances based on AMD's latest "Venice" CPUs for scenarios such as intelligent agent AI and semiconductor design. AMD claims that Helios integrates the company's self-developed GPUs, CPUs, networking chips, and software ecosystem, offering lower per-token costs. The system uses MI400 series GPUs, with each compute tray equipped with 4 Instinct GPUs and 1 EPYC CPU. Currently, Meta plans to deploy Helios systems at a scale of up to 1GW, and companies like OpenAI and Oracle have also announced adoption of related solutions. AMD stated that 8 out of the world's top 10 AI companies are already running workloads on its Instinct GPUs. [Original Link]