Artificial Analysis made its tests secret. The open-model gap went from 3 poin… · @swordBluesy · TwKit

@swordBluesy

Artificial Analysis made its tests secret. The open-model gap went from 3 points to 12. Best open-weight score: 32. Claude Opus 5.5: 58. Was China's catch-up real, or was the old test leaking? Free Open vs Closed Model Reality Check in the first reply.

曝光 13 · 评论 1 · 点赞 0 · 书签 0