gemini-4IMAGE
HLE 跑分:Gemini 4 Argon 57.1%,高于 Meta muse wonder 53.6%
Humanity's Last Exam 成绩单流出:Meta muse wonder 仅 53.6%,Gemini 4 Argon 57.1%,Claude Opus 5.5 以 61.4% 居首。
Sahil Panhotra | Indie Builder
@SahilPanhotra20 天前收录原推发布于 2026.10.13

Sahil Panhotra | Indie Builder @SahilPanhotra原推发布于 2026.10.13
muse wonder got cooked on humanity's last exam 😭 muse wonder full scored just 53.6% compared to 61.4% for opus 5.5 even gemini 4 argon managed to score 57.1% 💀 guess meta still has some catching up to do
这条是引用着另一条写的:x.com/sahilpanhotra/status/2109691435507495339
x.com/sahilpanhotra/status/2109693956133142672采集于 2026.10.13 09:59