UGI - Writing — leaderboard
Metric: Writing Score. Source: huggingface.co. 1234 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.5 Flash (Minimal) | 72.47 |
| 2 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | 70.7 |
| 3 | Claude Opus 4.5 (20251101) (Thinking) | 70.34 |
| 4 | Claude Opus 4.1 (Non-reasoning) | 69.91 |
| 5 | Gemini 3.5 Flash (High) | 69.71 |
| 6 | GPT-5.5 (xHigh) | 69.66 |
| 7 | GPT-5.4 (High) | 68.82 |
| 8 | GPT-5.4 (xHigh) | 68.65 |
| 9 | DeepSeek V4 Pro (Reasoning) | 68.4 |
| 10 | Claude Opus 4 (Non-reasoning) | 68.24 |
| 11 | Gemini 3.1 Pro (Preview) (High) | 68.22 |
| 12 | Claude 3.7 Sonnet (20250219) (Thinking) | 68.01 |
| 13 | GPT-5.5 (High) | 67.91 |
| 14 | Gemini 3 Flash (Preview) (Medium) | 67.71 |
| 15 | GPT-4.1 | 67.48 |
Interactive version: theaggregate.ai/benchmark?slug=ugi-writing · How the rankings work · Data refreshed daily, snapshot 2026-07-22.