WTF Is Going On(Weights, Tools & Frameworks)

한국어
NVIDIA filed

AI Infra Summit: NVIDIA Vera Rubin and DSX Platform Advancements Showcase Energy Efficiencies of Optimizing Tokens Per Watt for AI Factories

read it at the source — NVIDIA →

What happened

NVIDIA announced collaborations with partners including Amazon's Annapurna Labs and d-Matrix to optimize AI infrastructure. NVIDIA and its partners demonstrated new results, such as Lambda improving performance per watt by 23% with NVIDIA DSX MaxLPS. The NVIDIA Vera Rubin NVL72 is part of a full-stack AI factory platform that can deliver up to 1.4x more tokens per megawatt through factory-wide power optimization. NVIDIA DSX MaxLPS can enable up to 40% more GPU capacity within the same megawatt budget in certain deployment environments.

Our sentences, written from NVIDIA's own announcement and checked against it — every number and name must appear there.

Also on the board