SuperCLUE General (July 2025) - Math Reasoning: leaderboard
SuperCLUE general leaderboard, July 2025 edition (2025年7月): math reasoning (数学推理) score (总排行榜 sheet). SuperCLUE's periodic Chinese-language general evaluation uses a new question set each edition, so editions are separate boards. Score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 47 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 (High) | 81.6 | #132 (O3) |
| 2 | GPT-5 (High) | 76.86 | #105 (GPT-5) |
| 3 | O4 Mini (High) | 76.34 | #185 (O4 Mini) |
| 4 | DeepSeek R1 0528 | 75 | #232 |
| 5 | Gemini 2.5 Pro | 74.05 | #159 |
| 6 | DeepSeek V3.1 (Thinking) | 70.83 | #275 (DeepSeek V3.1) |
| 7 | Doubao-Seed-1.6-thinking-250715 | 70.77 | #169 |
| 8 | Grok 4 | 69.11 | #189 |
| 9 | Qwen 3 235B A22B 2507 (Thinking) | 68.7 | #263 |
| 10 | Hunyuan-T1-20250711 | 68.7 | #277 |
| 11 | GPT-OSS-120B | 67.18 | #346 |
| 12 | GLM-4.5 | 66.67 | #283 |
| 13 | Gemini 2.5 Flash | 65.89 | #255 |
| 14 | Qwen 3 235B A22B 2507 Instruct | 64.57 | #315 |
| 15 | Qwen 3 235B A22B (Thinking) | 63.36 | #323 (Qwen 3 235B A22B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-july-2025-math-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-05.