SuperCLUE General (July 2025) - Math Reasoning: leaderboard
Metric: Score. Source: www.superclueai.com. 47 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (High) | 81.6 |
| 2 | GPT-5 (High) | 76.86 |
| 3 | O4 Mini (High) | 76.34 |
| 4 | DeepSeek R1 0528 | 75 |
| 5 | Gemini 2.5 Pro | 74.05 |
| 6 | DeepSeek V3.1 (Thinking) | 70.83 |
| 7 | Doubao-Seed-1.6-thinking-250715 | 70.77 |
| 8 | Grok 4 | 69.11 |
| 9 | Qwen 3 235B A22B 2507 (Thinking) | 68.7 |
| 10 | Hunyuan-T1-20250711 | 68.7 |
| 11 | GPT-OSS-120B | 67.18 |
| 12 | GLM-4.5 | 66.67 |
| 13 | Gemini 2.5 Flash | 65.89 |
| 14 | Qwen 3 235B A22B 2507 Instruct | 64.57 |
| 15 | SenseNova V6 Reasoner | 63.36 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-july-2025-math-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.