Chinese AI model sets record with perfect score

Chinese technology company Xiaohongshu, or RedNote, said on Wednesday that its dots-note 3.0 model scored a perfect 42 points on this year's International Mathematical Olympiad problems, making it the first AI system in IMO history to achieve both a full score and a gold medal-level result.

The result makes dots-note 3.0 only the second AI model after Google DeepMind's Gemini Deep Think to receive an officially graded gold-level result. Gemini scored 35 points in 2025 after solving five of the six problems, matching that year's gold medal cutoff.

Xiaohongshu told China Daily that dots-note 3.0 used an agent-based reasoning system to generate its proofs, adding that the version used in the evaluation still remains internal and is expected to go open-source to the AI community in the near future.

Unlike many mathematical benchmarks that ask only for a final answer, the IMO requires complete proofs. A solution can lose points if any step is unsupported, making the contest a test of abstract reasoning as well as the consistency of a model's argument.

The six problems in question covered algebra, geometry, number theory and combinatorics, with each worth seven points. For the AI evaluation, human intervention was prohibited, including adjusting the model's reasoning, adding hints, correcting answers or selecting among outputs, according to the organizer.

The perfect score was rare even among human contestants. Seven of the more than 600 students listed in the official results received 42 points, or about 1 percent. The gold medal cutoff this year was 29 points, placing the model's reported result 13 points above the threshold.