Creative writing evaluation on EQ-Bench Creative
84.6S ScoreGPT-5.4
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5.42026.06 | 84.6 | |
| Claude Opus 4.62026.06 | 80.5 | |
| Gemini 3 Flash2026.06 | 75 | |
| Gemini 3.1 Pro2026.06 | 73.8 | |
| Gemma-4-31B2026.06 | 70.9 | |
| Gemma-4-26B-A4BMoE=true2026.06 | 70.5 | |
| POLARISBackbone=Qwen3.5-9B2026.06 | 70.3 | |
| Qwen3.5-27B2026.06 | 68.9 | |
| Gemma-4-E4BPLE=true2026.06 | 64.2 | |
| Qwen3.5-9B2026.06 | 59.2 | |
| Ministral-3-8B2026.06 | 53.9 | |
| DeepSeek-R1-Distill-Qwen-14B2026.06 | 42.8 | |
| LongWriter-Zero-32B2026.06 | 41.5 | |
| LongWriter-Llama3.1-8B2026.06 | 35.1 |