Math
MathArena
MathArena contest-math leaderboard (AIME, AMC, olympiad-style); results depend on answer-format extraction and evaluation date.
How models are tested
Scores are published numbers from the sources below. They are only comparable when the evaluation version, tools, agent harness, and sampling setup match. Missing scores are omitted rather than treated as zero.
- Category
- Math
- Metric
- Percent
- Direction
- Higher is better
- Coverage
- 4 / 318Models in this catalog with a published score
- Updated
- Catalog snapshot date
- Source
- matharena.ai/
Scores
- 1Claude Opus 584.4%
Anthropicclosed
- 2GPT-5.6 Sol79.7%
OpenAIclosed
- 3GPT-5.577.8%
OpenAIclosed
- 4Kimi K369.7%
Moonshotopen