SuperCLUE General (September 2025) - Agent: leaderboard
SuperCLUE general leaderboard, September 2025 edition (2025年9月): agent (智能体Agent) 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. 37 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 83.06 | #159 |
| 2 | Qwen 3 Max | 82.5 | #213 |
| 3 | Doubao-Seed-1.6-thinking-250715 | 82.5 | #169 |
| 4 | Gemini 2.5 Flash | 80.83 | #255 |
| 5 | Claude Sonnet 4.5 (Thinking) | 77.22 | #148 (Claude Sonnet 4.5) |
| 6 | GPT-5.1 (High) | 77.22 | #145 (GPT-5.1) |
| 7 | DeepSeek V3.2 Exp (Thinking) | 76.11 | #241 (DeepSeek V3.2 Exp) |
| 8 | DeepSeek V3.1 Terminus (Thinking) | 75.83 | #224 (DeepSeek V3.1 Terminus) |
| 9 | GPT-5 (High) | 75.28 | #105 (GPT-5) |
| 10 | Claude Opus 4.1 (Thinking) | 75.28 | #146 (Claude Opus 4.1) |
| 11 | Qwen 3 235B A22B 2507 (Thinking) | 74.44 | #263 |
| 12 | Gemini 3 Pro (Preview) | 73.89 | #71 |
| 13 | Grok 4 Fast (Reasoning) | 73.89 | #266 (Grok 4 Fast) |
| 14 | DeepSeek V3.1 Terminus | 71.94 | #224 |
| 15 | Grok 4 | 71.67 | #189 |
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-september-2025-agent · How It Works · Data refreshed daily, snapshot 2026-11-05.