# Meta AI 将其模型参加五项 STEM 奥林匹克竞赛，均获得金牌或满分成绩，证明其推理能力取得重大进展

- 来源：AI at Meta
- 发布时间：2026-08-06 23:34
- AIWatch 分数：66
- AIWatch 标记：未精选
- AIWatch 链接：https://aiwatch.icu/events/evt_01kzbvmfj8q9t6996qpj3hxmdv
- 原文链接：https://x.com/AIatMeta/status/2085388945148297322

## 精选理由

常规快讯，保留列表

## AI 摘要

Meta AI 将其模型参加五项 STEM 奥林匹克竞赛，均获得金牌或满分成绩，证明其推理能力取得重大进展。

## 正文

To understand whether we're making genuine progress on reasoning, we entered our AI models in five STEM Olympiad competitions.<br><br>The results:<br>🏅 Asian Physics Olympiad (APhO): Perfect score, theory exam<br>🏅 International Physics Olympiad (IPhO): Perfect score, theory exam<br>🥇 International Mathematical Olympiad (IMO): Gold medal<br>🥇 International Chemistry Olympiad (IChO): Gold-medal-level performance<br>🥇 Romanian Masters of Mathematics (RMM): Gold-medal-level performance<br><br>The types of problems in the Olympiad competitions are exceptionally hard, demanding deep chains of reasoning, creative insight, and flawless argumentation. To test pure reasoning capability, we disallowed all tool use, meaning no search, no coding, and no calculator.<br><br>We have deep admiration for the contestants and committees behind these competitions, and are grateful for their support in enabling our participation.<br><video width="1920" height="1080" src="https://video.twimg.com/amplify_video/2085386098293788673/vid/avc1/1920x1080/hEYfLtF8BHje4MRH.mp4?tag=29" controls="controls" poster="https://pbs.twimg.com/amplify_video_thumb/2085386098293788673/img/EgrnGpOKmuqyk4tM.jpg"></video>