ASUS在NVIDIA GTC 2026上公布了先进的液体冷却的AI系统,使高性能、高能效的AI计算成为可能。 ASUS unveiled advanced liquid-cooled AI systems at NVIDIA GTC 2026, enabling high-performance, energy-efficient AI computing.

ASUS在NVIDIA GTC 2026推出了新的液体冷却的AI基础设施,包括由NVIDIA的Vera Rubin平台驱动的架子系统和边缘服务器,每瓦可提供高达10x更高的性能和支持万亿参数模型。 ASUS launched new liquid-cooled AI infrastructure at NVIDIA GTC 2026, including rack-scale systems and edge servers powered by NVIDIA’s Vera Rubin platform, offering up to 10X higher performance per watt and support for trillion-parameter models. 该公司引进了混合冷却服务器、可扩缩的GPU连通性,以及AI公司房地部署的AI枢纽。 The company introduced hybrid-cooled servers, scalable GPU connectivity, and the AI Hub for on-premises AI deployment. Eaton, Schneider Electrical, Delta, and Trane Technologies 也推出综合电力、冷却和数字双解决方案, 以加速AI数据中心的建设, 提高效率, 支持GHAWAT规模的AI工厂。 Eaton, Schneider Electric, Delta, and Trane Technologies also unveiled integrated power, cooling, and digital twin solutions to accelerate AI data center construction, improve efficiency, and support gigawatt-scale AI factories. Nscale、微软、NVIDIA和Caterpillar宣布在西弗吉尼亚AI校园部署1.35GW GPU,并配有多兆瓦微电网,建造工作定于2027年开始。 Nscale, Microsoft, NVIDIA, and Caterpillar announced a 1.35GW GPU deployment at a West Virginia AI campus, backed by a multi-gigawatt microgrid, with construction set to begin in 2027.