OpenAI's Experimental Model Achieves Gold Medal Performance at International Math Olympiad 2025

Отредактировано: Veronika Radoslavskaya

OpenAI's experimental AI model has achieved gold medal-level performance at the 2025 International Math Olympiad (IMO), solving five out of six problems and earning 35 out of 42 points. This marks a significant advancement in AI's ability to tackle complex mathematical problems requiring creative reasoning.

The model was evaluated under the same conditions as human participants: two 4.5-hour sessions, no access to external tools or the internet, and required to write detailed proofs based on official IMO problems. Three former IMO medalists independently graded each solution, with final scores based on unanimous agreement.

Alexander Wei, a research scientist at OpenAI, emphasized the significance of this accomplishment, stating that the model can now "craft intricate, watertight arguments at the level of human mathematicians." This achievement underscores the rapid progress in AI's reasoning capabilities and its potential applications in various fields.

OpenAI does not plan to release this experimental model in the near term. The upcoming GPT-5, expected to be released soon, will likely be an improvement from its predecessor but won't feature the same level of mathematical prowess as the IMO-winning model.

Источники

  • Ars Technica

  • Simon Willison’s Weblog

  • CTOL Digital Solutions

  • Nice Math Problems

Вы нашли ошибку или неточность?

Мы учтем ваши комментарии как можно скорее.