WHAT NEW IN X11.02 MON

X 上的 AI 合集档案 · 22 个合集 · 64,204 条

gemini-4IMAGE

PLE 工程基准:Gemini 4 Argon 40.6% 落后 Opus 5.5

Physical-Engineering's Last Exam 榜单显示 Claude Opus 5.5 48.3%、GPT 6 Astra 45.8%、Gemini 4 Argon 40.6%、Grok 4.7 30.1%。

Yufan Zhuang
@yufan_zhuang8 天前收录原推发布于 2026.10.24
PLE 工程基准:Gemini 4 Argon 40.6% 落后 Opus 5.5
Yufan Zhuang @yufan_zhuang原推发布于 2026.10.24
在 X 上打开原推

Physical-Engineering’s Last Exam🛠️ Claude Opus 5.5: 48.3% GPT 6 Astra: 45.8% Gemini 4 Argon: 40.6% Grok 4.7: 30.1% Where AI Meets Real Engineering. 📄 Paper: https://arxiv.org/abs/2610.30537 🌐 Leaderboard: https://ple-bench.com 💻 Code: https://github.com/google-research/ple

x.com/yufan_zhuang/status/2113700290998518091采集于 2026.10.24 09:53