MathArena Apex: leaderboard
MathArena Apex is a math contest benchmark featuring the most difficult mathematical problems, testing advanced reasoning and problem-solving abilities of AI models. It focuses on olympiad-level mathematics and complex multi-step mathematical reasoning.
Metric: Score (self-reported). Source: matharena.ai. Status: years away from saturation. 48 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (xhigh) | 80.21 |
| 2 | Hy4 preview | 74.2 |
| 3 | GPT-5.4-Pro (xhigh) | 69.79 |
| 4 | Gemini 3.1 Pro Preview | 60.94 |
| 5 | GPT-5.4 (xhigh) | 54.17 |
| 6 | Qwen3.7 Max | 44.5 |
| 7 | Claude-Opus-4.7 (xhigh) | 40.62 |
| 8 | DeepSeek-V4-Pro | 38.3 |
| 9 | Nex-N2-Pro | 36.5 |
| 10 | Claude Opus 4.6 Max | 34.5 |
Interactive version: theaggregate.ai/benchmark?slug=matharena-apex · How It Works · Data refreshed daily, snapshot 2026-09-05.