SuperCLUE General (September 2025) - Precise Instruction Following: leaderboard
SuperCLUE general leaderboard, September 2025 edition (2025年9月): precise instruction following (精确指令遵循) 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: Around February 2027. 37 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.1 (High) | 73.53 | #145 (GPT-5.1) |
| 2 | GPT-5 (High) | 73.26 | #105 (GPT-5) |
| 3 | Claude Sonnet 4.5 (Thinking) | 69.25 | #148 (Claude Sonnet 4.5) |
| 4 | Gemini 3 Pro (Preview) | 68.45 | #71 |
| 5 | O3 (High) | 67.91 | #132 (O3) |
| 6 | O4 Mini (High) | 67.65 | #185 (O4 Mini) |
| 7 | Claude Opus 4.1 (Thinking) | 63.1 | #146 (Claude Opus 4.1) |
| 8 | Qwen 3 Next 80B A3B (Thinking) | 56.95 | #312 |
| 9 | DeepSeek V3.2 Exp (Thinking) | 55.61 | #241 (DeepSeek V3.2 Exp) |
| 10 | DeepSeek V3.1 Terminus (Thinking) | 54.81 | #224 (DeepSeek V3.1 Terminus) |
| 11 | Qwen 3 235B A22B 2507 (Thinking) | 51.6 | #263 |
| 12 | GPT-OSS-120B | 50 | #346 |
| 13 | Gemini 2.5 Pro | 49.73 | #159 |
| 14 | Kimi K2 (Thinking) | 48.93 | #252 (Kimi K2) |
| 15 | Hunyuan-T1-20250822 | 45.45 | #306 |
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-precise-instruction-following · How It Works · Data refreshed daily, snapshot 2026-11-05.