AI 热点

Gemini 4 Hydrogen推理档位成绩分化

ARC Prize更新于 11月21日 01:29

@GoogleDeepMind On ARC-AGI-2, Gemini 4 Hydrogen scored higher than Gemini 3.8 Flash at medium (89.2% vs 82.9%) and high (90.0% vs 89.2%), though trailed at low

AI 导读

在 ARC-AGI-2 上,Gemini 4 Hydrogen 在 medium 和 high 推理档位分别以 89.2% 和 90.0% 超过 Gemini 3.8 Flash 的 82.9% 和 89.2%,但在 low 档位以 65.8% 落后于后者的 77.5%。Gemini 4 Hydrogen 还新增了 Gemini 3.8 Flash 没有的 xhigh 推理档位。

正文

@GoogleDeepMind On ARC-AGI-2, Gemini 4 Hydrogen scored higher than Gemini 3.8 Flash at medium (89.2% vs 82.9%) and high (90.0% vs 89.2%), though trailed at low reasoning (65.8% vs 77.5%).

Gemini 4 Hydrogen also introduced a xhigh reasoning level that was not available for Gemini 3.8 Flash. https://t.co/yEewYk1IuM

来源

ARC Prize