Recent official testing of IMO 2026 problems has shown select frontier models, including Huawei’s Celia and Xiaohongshu’s dots-note-3.0, achieving perfect 42/42 scores when submitted through the competition’s formal post-contest channel and graded by organizers. These results follow rapid capability gains in large language models on complex proof-based reasoning benchmarks. Trader sentiment still tilts slightly toward “No” at 57 percent because independent verification of full contest conditions, absence of human intervention, and precise market-resolution thresholds remain contested, while self-administered tests on other models have drawn skepticism. Key upcoming catalysts include any final IMO statements on AI eligibility and clearer documentation of submission protocols that could shift the aggregated capital-weighted odds.