Creative Writing v3 Elo
Chat & Writing · 2025-03-28
This Creative Writing v3 metric ranks model responses through pairwise judgments on creative-writing prompts. Scores are Elo ratings, with higher values indicating stronger writing.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 2430 |
| Kimi K3 | 2340 |
| GPT-5.6 Sol | 2092 |
| Claude Fable 5 | 2065 |
| Opus 4.7 | 2032 |
| Muse Spark 1.1 | 2004 |
| GPT-5.5 | 1908 |
| GPT-5.6 Luna | 1899 |
| Opus 4.8 | 1889 |
| GPT-5.6 Terra | 1887 |
| GPT-5.4 | 1886 |
| Opus 4.6 | 1874 |
| Sonnet 4.6 | 1873 |
| Sonnet 5 | 1782 |
| Sonnet 4.5 | 1732 |
| o3 | 1728 |
| GLM-5.2 | 1720 |
| GPT-5.2 | 1718 |
| Opus 4.5 | 1715 |
| GPT-5.3 Instant | 1713 |
| Kimi K2.6 | 1710 |
| GPT-5.4 Mini | 1678 |
| Kimi K2 Instruct | 1669 |
| Kimi K2 Thinking | 1666 |
| GPT-5 (2025-08-07) | 1648 |
| Grok 4.20 | 1634 |
| GLM-5 | 1626 |
| Inkling | 1614 |
| Horizon Alpha | 1613 |
| GLM-5.1 | 1605 |
| Grok 4.5 | 1581 |
| Kimi K2.5 | 1577 |
| Opus 4 | 1573 |
| DeepSeek-V4-Pro | 1569 |
| DeepSeek-V4-Flash | 1554 |
| DeepSeek-R1 | 1500 |
| Gemini 3 Pro Preview | 1491 |
| DeepSeek-V3.2 | 1489 |
| Mistral Medium 3.1 | 1485 |
| MiMo-V2.5-Pro | 1482 |