SuperCLUE General (March 2026) - Precise Instruction Following: leaderboard
Metric: Score. Source: www.superclueai.com. 24 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) (High) | 56.76 |
| 2 | Claude Opus 4.6 (Max) | 47.57 |
| 3 | GPT-5.4 (xHigh) | 44.32 |
| 4 | DeepSeek V4 Pro (Max) | 37.84 |
| 5 | Gemini 3 Flash (Preview) (High) | 35.68 |
| 6 | DeepSeek V4 Flash (Max) | 32.43 |
| 7 | Grok 4.20 Beta (0309) (Reasoning) | 32.43 |
| 8 | DeepSeek V3.2 (Thinking) | 25.95 |
| 9 | GLM-5 | 24.86 |
| 10 | GPT-OSS-120B (High) | 20.54 |
| 11 | Qwen 3.5 397B A17B (Thinking) | 19.46 |
| 12 | MiniMax-M2.7 | 17.3 |
| 13 | Kimi K2.5 (Thinking) | 16.22 |
| 14 | MiMo-V2-Pro | 16.22 |
| 15 | Tencent HY 2.0 Think | 14.05 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-general-march-2026-precise-instruction-following · How It Works · Data refreshed daily, snapshot 2026-09-19.