Benchmarks · creative
UGI Writing
A maintainer-owned writing evaluation considering intelligence, style, repetition and length adherence, informed by human preferences. Maintained by DontPlanToEnd. Score: Writing ✍️.
166 models · 167 results · data as of 2026-10-29 · huggingface.co ↗
Leaderboard
Gemini 4 Argon max
All results
167 rows| # | Model | Effort | UGI Writing |
|---|---|---|---|
| 1 | Gemini 4 Argon | max | 79.7 |
| 2 | Gemini 3.7 Flash | max | 76.1 |
| 3 | Gemini 3.8 Flash | max | 72.7 |
| 4 | Gemini 3.5 Flash | xhigh | 72.5 |
| 5 | Gemini 3.1 Pro | high | 72.2 |
| 6 | Fable 5 | max | 71.9 |
| 7 | Opus 4.5 | high | 70.4 |
| 8 | Opus 5.5 | max | 70.0 |
| 9 | Opus 4 | medium | 68.2 |
| 10 | GPT-4.1 | high | 67.5 |
| 11 | Grok 4.0709 | 67.0 | |
| 12 | Sonnet 4.5 | high | 66.9 |
| 13 | Opus 4.1 | high | 66.5 |
| 14 | Opus 4.6 | max | 66.1 |
| 15 | o3 | high | 65.8 |
| 16 | Gemini 3.6 Flash | xhigh | 65.7 |
| 17 | Gemini 3 Pro | high | 65.6 |
| 18 | Gemini 2.5 Pro | high | 65.2 |
| 19 | GPT-6 Astra | max | 65.1 |
| 20 | Gemini 3 Flash | high | 65.0 |
| 21 | Opus 4.8 | max | 64.7 |
| 22 | Kimi K2.6 | max | 64.2 |
| 23 | Sonnet 4 | high | 63.7 |
| 24 | GPT-4o | high | 63.4 |
| 25 | Grok 4.6 | max | 63.3 |
| 26 | Grok 4.20 | xhigh | 63.1 |
| 27 | GPT-5 | high | 62.7 |
| 28 | Kimi K2.5 | max | 61.5 |
| 29 | GLM 5.2 | max | 61.2 |
| 30 | o1 | high | 61.0 |