Role-Play Generation on Role-Play Generation Ability Evaluation
4.406DFDeepSeek-V3.1
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DeepSeek-V3.12025.12 | 4.406 | 4.564 | 4.55 | 4.912 | 4.55 | 4.596 | |
| Claude-3.5-sonnet2025.12 | 4.26 | 4.433 | 3.974 | 4.38 | 4.015 | 4.174 | |
| Claude-3.5-haiku2025.12 | 4.155 | 3.778 | 4.027 | 4.167 | 3.556 | 3.937 | |
| GPT-4o-Mini2025.12 | 3.926 | 4.185 | 4.278 | 3.889 | 3.63 | 3.982 | |
| Qwen2.5-72B-Instructparameters=72B2025.12 | 3.91 | 4.232 | 4.253 | 4.56 | 4.106 | 4.212 | |
| Doubao-1.5-lite2025.12 | 3.892 | 4.42 | 4.295 | 3.994 | 4.127 | 4.146 | |
| Qwen2.5-14B-Instructparameters=14B2025.12 | 3.333 | 4.132 | 4.028 | 4.333 | 3.557 | 3.877 | |
| Baichuan2-13B-Chatparameters=13B2025.12 | 3.25 | 4.017 | 3.25 | 3.584 | 3.047 | 3.425 | |
| Llama-3.1-70B-Instructparameters=70B2025.12 | 3.142 | 2.542 | 3.176 | 3.312 | 3.68 | 3.162 | |
| chatglm3-6bparameters=6b2025.12 | 2.35 | 2.507 | 2.412 | 3 | 2.527 | 2.559 | |
| Qwen2.5-7B-Instructparameters=7B2025.12 | 2.238 | 2.191 | 2.512 | 3.071 | 2.333 | 2.469 | |
| Llama-3.1-8Bparameters=8B2025.12 | 2.222 | 2.667 | 2.5 | 2.08 | 2.567 | 2.407 | |
| Hunyuan-7B-Instructparameters=7B2025.12 | 2.2 | 2.333 | 2.253 | 2.7 | 3.313 | 2.56 |