AI 导读
Astra 在 Mystery Game Puzzles 上得分非常高，原始分数为 84.0%。第二高的分数是 59.0%，由 Claude Opus 5 取得。MGP 用一款知名游戏的解谜变体中的 100 个局面来测试模型，该游戏的细节被刻意保密。 https://t.co/gl0S7cXwVh
正文
Astra scored very high on Mystery Game Puzzles, with a raw score of 84.0%. The second-highest score is 59.0%, set by Claude Opus 5. MGP tests models on 100 positions from a puzzle-oriented variant of a well-known game, the details of which are deliberately left undisclosed.
https://t.co/gl0S7cXwVh
来源：@EpochAIResearch · x.com