Two Chinese AI systems were reported to score a perfect 42 of 42 at the 2026 Maths Olympiad
AI from Xiaohongshu and Huawei was reported as officially graded 42 of 42. Seven of 666 students also got full marks. In 2024 AI scored 28.
Best AI score at the International Mathematical Olympiad
Points out of 42 on the year's IMO paper, for results graded by or reported as graded by IMO coordinators.
What this measures: The International Mathematical Olympiad is the world's top math contest for high-school students: six proof problems, 42 points in total. Gold usually needs about 29–35.
- Best AI score
- Reference level
Saturated: AI systems were reported to reach a perfect 42/42 in 2026, so the contest can't show further progress. The next frontier is open research problems. Now watch: FrontierMath Erdős and Erdős problems solved.
Gold-medal cutoffs: 35 (2025), 29 (2026). The 2026 results were reported by Chinese media and the companies; we found no statement published by the IMO itself.
Show data table and sources
| Date | Label | Value |
|---|---|---|
| 25 Jul 2024 | AlphaProof + AlphaGeometry 2 | 28 points |
| 21 Jul 2025 | Gemini Deep Think | 35 points |
| 22 Jul 2026 | dots-note-3.0 / Huawei Celia | 42 points |
| Perfect score | 42 points |
- Advanced version of Gemini with Deep Think officially achieves gold-medal standard (Google DeepMind)
- AI catches up with humans to score 100% at top maths contest (France 24 / AFP)
- dots-note-3.0 IMO 2026 perfect score (Chinese) (China Daily)
The 2026 International Mathematical Olympiad (IMO), a maths contest for school students, was held in Shanghai on 15–16 July. Gold needed 29 of 42 points, against 35 in 2025, which points to a harder paper.
Xiaohongshu (RedNote) says its AI system dots-note-3.0 solved all six problems for 42/42. According to China Daily, the IMO organising committee graded the proofs. Labs got each day’s problems only after the students finished, had to submit by a deadline and could not get any human help. Huawei announced the same score for its Celia (Xiaoyi) assistant. A Chinese outlet quotes IMO President Gregor Dolinar congratulating Huawei on a perfect 42.
Seven of 666 human contestants also scored 42, according to AFP. The best officially graded AI score has gone from 28/42 (2024) to 35/42 (2025) to 42/42 (2026).
What it means
If the reports hold, AI now matches the very top students at this contest, not just the gold-medal line. These are hard problems with known solutions, not open research, and the grading claims rest on media reports and company statements rather than anything the IMO itself has published.
Caveats
- We found no statement published by the IMO itself about either AI result. The grading claims come from Chinese media reports and company announcements.
- No computing budget, number of attempts or full method was disclosed.
- Other labs’ models were reported to score 42/42 in unofficial runs after the contest. The IMO did not grade those, so they aren’t counted here.
- Reported contestant numbers differ between outlets (666 vs about 685).