SuperCLUE General (September 2025) - Math Reasoning: leaderboard
Metric: Score. Source: www.superclueai.com. 37 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.1 (High) | 82.73 |
| 2 | Kimi K2 (Thinking) | 80.91 |
| 3 | Gemini 3 Pro (Preview) | 79.09 |
| 4 | O3 (High) | 77.27 |
| 5 | GPT-5 (High) | 73.64 |
| 6 | O4 Mini (High) | 72.73 |
| 7 | Gemini 2.5 Pro | 71.82 |
| 8 | GPT-OSS-120B | 70.91 |
| 9 | Qwen 3 Next 80B A3B (Thinking) | 66.36 |
| 10 | Gemini 2.5 Flash | 62.73 |
| 11 | Qwen 3 Max | 62.73 |
| 12 | Grok 4 Fast (Reasoning) | 62.73 |
| 13 | DeepSeek V3.2 Exp (Thinking) | 60.91 |
| 14 | Doubao-Seed-1.6-thinking-250715 | 60.55 |
| 15 | DeepSeek V3.1 Terminus (Thinking) | 60 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-september-2025-math-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.