面向未来 Vera Rubin 平台的容量与吞吐量提升仍属英伟达预测;Lambda 的数据则是基于 Blackwell 系统的已披露部署验证结果。
What did Nvidia present on the opening day of its AI Infra Summit about transforming AI factories from systems optimized primarily for raw pNvidia’s AI Infra Summit focused on increasing useful AI output within fixed power budgets.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What did Nvidia present on the opening day of its AI Infra Summit about transforming AI factories from systems optimized primarily for raw p. Article summary: Nvidia’s opening-day message was that an AI factory should be measured not only by peak compute, but by validated agentic tokens produced per megawatt—requiring integrated design across silicon, systems, networking, soft. Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fa
openai.com
AI 基础设施面临的约束,正越来越多地来自可获得的电力,而不只是机架里还能塞下多少块加速器。英伟达在 2026 年 AI Infra Summit 首日提出:衡量 AI 工厂产出的核心指标,应是每兆瓦可产出的智能体 Token 数——即在既定供电上限内,一座设施究竟能完成多少有用的推理工作。23
英伟达超大规模与高性能计算副总裁 Ian Buck 在美国加州圣克拉拉举行的活动上阐述了这一思路。重点并不止于 GPU 跑得多快,而是要把计算、网络、推理软件、散热、电源控制与电网运营协同起来;当用电成为硬约束时,AI 工厂才能持续提升有效产出。123
在英伟达 Eos AI 工厂,Emerald AI 的 Conductor 平台与 Silicon Valley Power 合作,参与了商用柔性负载项目。英伟达称,该系统响应了超过 200 次来自电力公司的需求信号;在一次演示中,设施功耗在不到一分钟内从 4 MW 降至 3 MW。实现方式是降低可灵活调度计算任务的优先级,同时让优先级更高的推理服务继续运行。23