← Back
21 July 2025Benchmark
Google DeepMind and OpenAI models reach gold-medal level at the IMO
An advanced Gemini Deep Think became the first AI officially graded at gold-medal standard at the 2025 International Mathematical Olympiad (35/42), days after OpenAI reported a similar self-graded result from an experimental reasoning model.
Actors involved
Tags
imomathreasoningdeep-think