通过更智能的硬件和软件, 减少人工智能能源使用量并提高性能35倍. Nvidia's 2026 Vera Rubin platform slashes AI energy use and boosts performance 35x via smarter hardware and software.

恩维迪亚的Vera Rubin平台于2026年推出,它提高了人工智能效率,性能比前者提升了35倍,使用动态电源管理和集成硬件在现有功率限制内提供30%以上的能力. Nvidia’s Vera Rubin platform, launched in 2026, boosts AI efficiency with a 35x performance leap over its predecessor, using dynamic power management and integrated hardware to deliver 30% more capacity within existing power limits. 它是为代理人工智能设计的,它统一计算,存储和网络,使得部署速度更快,利用率更高. Designed for agentic AI, it unifies compute, storage, and networking, enabling faster deployment and higher utilization. 一个新的蓝色场-4 STX参考设计通过嵌入处理在存储中来增强数据处理. A new BlueField-4 STX reference design enhances data handling by embedding processing in storage. 与此同时,新兴的人工智能模型通过优化架构和算法将能源消耗降低高达70%,有助于制数据中心不断增长的电力需求,2024年该数据中心使用了415兆瓦时 (超过10%美国电力的用量). Meanwhile, emerging AI models are cutting energy use by up to 70% through optimized architecture and algorithms, helping curb the rising power demands of data centers, which consumed 415 terawatt-hours in 2024—over 10% of U.S. electricity. 这些进展标志着向更智能,更可持续的AI基础设施转变. These advances signal a shift toward smarter, more sustainable AI infrastructure.