SuperCLUE-Writing - Constrained Writing: leaderboard
Metric: Score. Source: www.superclueai.com. 14 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 (xHigh) | 96.17 |
| 2 | Claude Opus 4.7 (Max) | 95.41 |
| 3 | Gemini 3.1 Pro (Preview) | 93.94 |
| 4 | Grok 4.20 0309 (Reasoning) | 93 |
| 5 | GLM-5.1 | 92.53 |
| 6 | Doubao-Seed-2.0-Pro-260215 | 91.03 |
| 7 | Kimi K2.6 (Thinking) | 86.05 |
| 8 | Ernie 5.1 | 84.47 |
| 9 | MiniMax-M2.7 | 76.72 |
| 10 | DeepSeek V4 Pro (Max) | 74.62 |
| 11 | DeepSeek V4 Flash (Max) | 65.16 |
Interactive version: theaggregate.ai/benchmark?slug=superclue-writing-constrained-writing · How It Works · Data refreshed daily, snapshot 2026-09-05.