SuperCLUE General (July 2025) - Precise Instruction Following: leaderboard
SuperCLUE general leaderboard, July 2025 edition (2025年7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: Estimated already saturated. 47 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 (High) | 66.77 | #105 (GPT-5) |
| 2 | O3 (High) | 64.09 | #132 (O3) |
| 3 | O4 Mini (High) | 59.64 | #185 (O4 Mini) |
| 4 | DeepSeek V3.1 (Thinking) | 47.18 | #275 (DeepSeek V3.1) |
| 5 | Claude Opus 4 (Thinking) | 44.81 | #170 (Claude Opus 4) |
| 6 | Qwen 3 235B A22B 2507 (Thinking) | 43.62 | #263 |
| 7 | Hunyuan-T1-20250711 | 42.73 | #277 |
| 8 | GPT-OSS-120B | 41.25 | #346 |
| 9 | Gemini 2.5 Pro | 39.76 | #159 |
| 10 | GPT-OSS-20B | 35.31 | #523 |
| 11 | DeepSeek R1 0528 | 35.31 | #232 |
| 12 | Gemini 2.5 Flash | 34.12 | #255 |
| 13 | Doubao-Seed-1.6-thinking-250715 | 32.34 | #169 |
| 14 | Grok 4 | 29.97 | #189 |
| 15 | 360Zhinao2-O1.5 | 28.49 | #395 |
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-precise-instruction-following · How It Works · Data refreshed daily, snapshot 2026-11-05.