Kimi K2.7 ranks second behind Fable 5 and above GPT 5.5 xhigh in ErdosBench's mathematical research test
Przemek Chojecki's 14-problem smoke run puts Moonshot's new open-weight model behind Claude Fable-5-max and ahead of GPT-5.5 xhigh.
点赞
收藏
点踩
分享评论
?
参与讨论
Przemek Chojecki's 14-problem smoke run puts Moonshot's new open-weight model behind Claude Fable-5-max and ahead of GPT-5.5 xhigh.