@SemiAnalysis_: At Stanford CS153 Frontier Systems, Jensen states word for word that he "would like to be at low MFU all the time" & the reasoning Jensen gi...

@SemiAnalysis_ 3 信息等级 3 发布：2026-05-17T02:20 抓取：2026-05-17 04:02

🔗 原文链接

AI 算力行业

摘要

黄仁勋在斯坦福大学演讲中表示，他希望始终维持低模型浮点利用率（MFU），通过过度配置算力、网络和内存等资源来实现更高智能，并暗示xAI可能遵循此策略。

客观事实

黄仁勋称希望始终处于低MFU状态。
原因是通过过度配置算力、网络和内存来提升智能。
黄仁勋暗示xAI可能遵循此哲学。

黄仁勋斯坦福大学 xAI

原文

At Stanford CS153 Frontier Systems, Jensen states word for word that he "would like to be at low MFU all the time" & the reasoning Jensen gives is that he wants be so smart, he over-provisioned the work like flops, networking, memory, etc. maybe the kernel folks at @xai are following this philosophy too

likes: 53 | retweets: 1 | replies: 7 | views: 7824