[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"aihot-art-99736":3},{"itemId":4,"vertical":5,"category":6,"source":7,"score":8,"title":9,"summary":10,"analysis":11,"url":12,"coverUrl":13,"direction":13,"marketSignal":13,"publishedAt":14},"99736","ai","行业动态","Berry Xia",60,"多款AI模型在IMO 2026中获满分","GPT-SOL-5.6、Qwen3.8-Max 和 Kimi K3 在 IMO 2026 中均获满分，Claude-Fable-5 未完成。","多款AI模型在IMO 2026中获满分，GPT-SOL-5.6、Qwen3.8-Max和Kimi K3均以42分完成，Claude-Fable-5失败。\n· 三个模型一次性解决所有6道题目，满分42分，速度差异明显。\n· 去年仅有少数模型接近满分，今年多个主流模型已能稳定满分，显示AI推理能力飞跃。\n· Claude-Fable-5未能完成任务，表明不同模型在数学推理上仍有差距。\n· IMO成绩成为衡量AI推理能力的重要基准，满分突破具有里程碑意义。\n看点：AI在IMO上的表现将推动数学教育、科研辅助等领域的应用变革。","https:\u002F\u002Fx.com\u002Fberryxia\u002Fstatus\u002F2079583917045317659",null,"2026-07-21 23:06:57"]