LLM Creative Story Writing
Chat & Writing · 2025-01-05
Independent creative-writing benchmark comparing short stories generated by LLMs. The current leaderboard reports model scores and win probabilities from pairwise comparisons.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5.1 | 3.8 |
| Claude Opus 5 | 3.7 |
| Claude Opus 5.5 | 3.7 |
| GPT-6 Astra | 3.4 |
| GLM-5.3 | 3.0 |
| Claude Fable 5 | 2.9 |
| GPT-5.6 Sol | 2.5 |
| GPT-5.5 | 2.4 |
| Kimi K3 | 2.4 |
| GPT-5.4 | 2.0 |
| Muse Wonder | 2.0 |
| Opus 4.7 | 1.8 |
| Sonnet 4.6 | 1.7 |
| MiMo-V2.6-Pro | 1.6 |
| Opus 4.6 | 1.1 |
| Gemini 4 Argon | 0.8 |
| Opus 4.8 | 0.8 |
| Muse Spark 1.1 | 0.8 |
| Muse Spark 1.3 | 0.6 |
| GLM-5.2 | 0.5 |
| Grok 4.7 | 0.4 |
| GPT-5.2 | 0.3 |
| Gemini 3.8 Flash | 0.1 |
| Kimi K2.6 | 0.1 |
| Muse Spark 1.2 | -0.0 |
| MiniMax M3 | -0.1 |
| Mistral Medium 3.1 | -0.4 |
| DeepSeek V4.1 Flash | -0.5 |
| Qwen3.8 27B | -0.6 |
| MiMo-V2.5-Pro | -0.7 |
| Qwen3 Max (rolling alias) | -0.7 |
| Gemini 3.7 Flash | -0.7 |
| Qwen3.6-Max-Preview | -0.9 |
| GLM-5.1 | -1.0 |
| Kimi K2.5 | -1.1 |
| MiMo-V2-Pro | -1.2 |
| ERNIE 5.1 | -1.3 |
| Mistral Large 4 Preview | -1.6 |
| Mistral Large 3 675B Instruct 2512 | -1.8 |
| Gemma 4 31B IT | -1.9 |