gemini-4IMAGE

Gemini 4 Argon 数学能力实测：FrontierMath Tier 4 仅 78%

用户实测 Gemini 4 Argon 在 FrontierMath Tier 4 得 78%，试解 Erdős 问题自称有进展但大多错误，认为它难以用于科研数学。

Leeham

@Liam0697245222 天前收录原推发布于 2026.10.15

Gemini 4 Argon 数学能力实测：FrontierMath Tier 4 仅 78%

Leeham @Liam06972452原推发布于 2026.10.15

Gemini 4 Argon scores 78% on Frontiermath Tier 4 which isn't great. I tried a random Erdos problem with the model in which it claimed to make substantial progress, however it was mostly incorrect. I don't think this model can realistically be used for research math.

x.com/liam06972452/status/2110507498977550365采集于 2026.10.15 11:25