Noometry

Writing & Preference benchmark

Short-Story Creative Writing leaderboard

As of October 2026, GPT-5 has the highest published Short-Story Creative Writing score on Noometry at 86%, out of 39 models with results.

Last verified

About Short-Story Creative Writing

Short stories that must work in required elements, graded by a panel of LLM judges on a 10-point scale (shown here out of 100).

Category
Writing & Preference
Introduced
2025
Format
Graded writing
Unit
Percent (random guessing ≈ 0%)
Official site
github.com

Top 15 models

  1. GPT-5 86%
  2. Kimi K2 (Jul 2025) 85.6%
  3. Claude Opus 4.1 84.7%
  4. o3-pro 84.4%
  5. o3 83.9%
  6. Gemini 2.5 Pro 83.8%
  7. Claude Opus 4 83.6%
  8. GPT-5 Mini 83.1%
  9. DeepSeek-R1 83%
  10. Qwen3 235B-A22B 83%
  11. GPT-4o 81.8%
  12. Claude Sonnet 4 81.4%
  13. Claude 3.7 Sonnet 81.1%
  14. Claude 3.5 Sonnet 80.3%
  15. QwQ-32B 80.2%

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

#ModelProviderScoreSettingSourceDate
1GPT-5 OpenAI86%mediumEpoch AI
2Kimi K2 (Jul 2025) Moonshot AI85.6%Epoch AI
3Claude Opus 4.1 Anthropic84.7%Epoch AI
4o3-pro OpenAI84.4%mediumEpoch AI
5o3 OpenAI83.9%mediumEpoch AI
6Gemini 2.5 Pro Google83.8%Epoch AI
7Claude Opus 4 Anthropic83.6%16KEpoch AI
8GPT-5 Mini OpenAI83.1%mediumEpoch AI
9DeepSeek-R1 DeepSeek83%Epoch AI
10Qwen3 235B-A22B Alibaba (Qwen)83%Epoch AI
11GPT-4o OpenAI81.8%Epoch AI
12Claude Sonnet 4 Anthropic81.4%16KEpoch AI
13Claude 3.7 Sonnet Anthropic81.1%16KEpoch AI
14Claude 3.5 Sonnet Anthropic80.3%Epoch AI
15QwQ-32B Alibaba (Qwen)80.2%Epoch AI
16Gemma 3 27B Google79.9%Epoch AI
17Mistral Medium Mistral AI77.3%Epoch AI
18gpt-oss-120b OpenAI77.1%Epoch AI
19DeepSeek-V3 DeepSeek77%Epoch AI
20Grok 4 xAI76.9%Epoch AI
21Gemini 2.5 Flash Google76.5%Epoch AI
22Grok 3 xAI76.4%Epoch AI
23GPT-4.5 OpenAI75.6%Epoch AI
24Qwen3-30B-A3B Alibaba (Qwen)75.3%Epoch AI
25o4-mini OpenAI75%mediumEpoch AI
26Gemini 2.0 Flash (Feb 2025) Google73.8%Epoch AI
27Claude 3.5 Haiku Anthropic73.5%Epoch AI
28Grok-3 mini xAI73.5%lowEpoch AI
29GLM-4.5 Z.ai (Zhipu)73.4%Epoch AI
30Qwen2.5-Max Alibaba (Qwen)72.9%Epoch AI
31o1 OpenAI70.2%mediumEpoch AI
32Mistral Large Mistral AI69%Epoch AI
33GPT-4o mini OpenAI67.2%Epoch AI
34o1-mini OpenAI64.9%mediumEpoch AI
35Grok-2 (Dec 2024) xAI63.6%Epoch AI
36Phi-4 Microsoft62.6%Epoch AI
37Llama 4 Maverick Meta62%Epoch AI
38o3-mini OpenAI61.7%highEpoch AI
39Amazon Nova Pro Amazon60.5%Epoch AI

Compare the leaders

Other writing & preference benchmarks

Frequently asked questions

What does Short-Story Creative Writing measure?

Short stories that must work in required elements, graded by a panel of LLM judges on a 10-point scale (shown here out of 100).

Which model has the highest Short-Story Creative Writing score?

As of October 2026, GPT-5 has the highest published Short-Story Creative Writing score on Noometry at 86%, out of 39 models with results.

What is the best open-weight model on Short-Story Creative Writing?

Kimi K2 (Jul 2025) has the highest Short-Story Creative Writing accuracy among open-weight models at 85.6%, ranking 2 of 39 overall.