An advanced version of Google DeepMind's Gemini model, utilizing its "Deep Think" mode, has achieved a gold-medal standard at the International Mathematical Olympiad (IMO). The model successfully solved five out of six problems, earning 35 points, a significant improvement from last year's silver-medal performance. Notably, this year's Gemini operated end-to-end in natural language, generating proofs directly from problem descriptions within the competition's time limit, unlike previous systems that required translation into formal languages. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
RANK_REASON Frontier-lab model release with system card
