USAMO 2026

Grades full written proofs, not just final answers, for the six 2026 USA Mathematical Olympiad problems.

Also known as: MathArena USAMO 2026

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorymath
Subcategoryolympiad proof-writing
Page statusactive
Metric% of maximum rubric points (proof grading)
Directionhigher_is_better
Unit%
Dataset size6
Dataset licenceCC BY-NC-SA 4.0
PublisherSRI Lab, ETH Zurich, with INSAIT (MathArena project); evaluation partly supported by a Google grant

What it measures

USAMO 2026 evaluates whether a model can produce a complete, rigorous mathematical proof, not just a final numeric answer, for the six problems of the 2026 USA Mathematical Olympiad. USAMO is a proof-based competition, so each problem asks for a full written argument, testing multi-step mathematical reasoning and the ability to communicate a valid proof rather than final-answer pattern matching.

Task format

six open-ended proof problems in LaTeX; a model produces a full written solution, graded against a rubric

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
Claude Mythos PreviewAnthropic97.62026-04
GPT-5.4OpenAI95.22026-04
GLM 5.1Z.ai (Zhipu AI)83.82026-04
Gemini 3.1 Pro PreviewGoogle DeepMind74.42026-04
Claude Opus 4.6Anthropic42.32026-04

Data

This page as JSON · Edit on GitHub