Character Roleplay on ARCANE validated slice (5 novels) (In-Scenario)
69.5APFDeepSeek-V4
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| DeepSeek-V4Size=Pro, Mode=Arc2026.06 | 69.5 | 67.7 | 62.8 | 65.1 | |
| DeepSeek-V4Size=Pro, Mode=LifeChoice2026.06 | 69.1 | 66.9 | 63 | 64 | |
| DeepSeek-V4Size=Pro, Mode=RAG2026.06 | 68.1 | 66.1 | 61.8 | 63 | |
| DeepSeek-V4Size=Pro, Mode=Summary2026.06 | 67.1 | 65.6 | 60.2 | 61.4 | |
| DeepSeek-V4Size=Flash, Mode=Arc2026.06 | 66.8 | 65.8 | 59.2 | 62.1 | |
| DeepSeek-V4Size=Flash, Mode=RAG2026.06 | 66.3 | 64.7 | 59.1 | 60.9 | |
| DeepSeek-V4Size=Flash, Mode=LifeChoice2026.06 | 66.3 | 64.7 | 59.1 | 61.7 | |
| ARCANESize=32B, Mode=RAG2026.06 | 65.8 | 63.5 | 59.2 | 55.8 | |
| DeepSeek-V4Size=Flash, Mode=Summary2026.06 | 64.1 | 63 | 56.4 | 59.6 | |
| ARCANESize=32B, Mode=LifeChoice2026.06 | 62.1 | 60.2 | 54.6 | 53.6 | |
| ARCANESize=8B, Mode=RAG2026.06 | 61.4 | 59.5 | 54.4 | 53 | |
| DeepSeek-V4Size=Pro, Mode=Vanilla2026.06 | 61.3 | 60.2 | 53.4 | 56.1 | |
| DeepSeek-V4Size=Flash, Mode=TimeChara2026.06 | 60.8 | 60.2 | 53.2 | 55.6 | |
| DeepSeek-V4Size=Pro, Mode=TimeChara2026.06 | 60.4 | 59.5 | 53.1 | 54.8 | |
| DeepSeek-V4Size=Flash, Mode=Vanilla2026.06 | 59.8 | 59.3 | 52.1 | 53.8 | |
| ARCANESize=8B, Mode=Arc2026.06 | 59.5 | 57.7 | 50.6 | 56 | |
| Qwen3Size=32B, Mode=RAG2026.06 | 59.4 | 57.3 | 50.8 | 49.2 | |
| ARCANESize=32B, Mode=Arc2026.06 | 59.4 | 58 | 50.9 | 53.4 | |
| Qwen3Size=32B, Mode=LifeChoice2026.06 | 58.3 | 56.6 | 49.1 | 50.5 | |
| ARCANESize=8B, Mode=LifeChoice2026.06 | 58.2 | 56.2 | 50.1 | 50.5 | |
| Qwen3Size=32B, Mode=Arc2026.06 | 57.4 | 55.6 | 47.4 | 51.6 | |
| ARCANESize=32B, Mode=Summary2026.06 | 57 | 55.8 | 48.4 | 49.5 | |
| ARCANESize=8B, Mode=Summary2026.06 | 56.5 | 55 | 47.7 | 49.7 | |
| Qwen3Size=32B, Mode=Summary2026.06 | 54.8 | 53.6 | 45.2 | 46.7 | |
| ARCANESize=32B, Mode=TimeChara2026.06 | 53.7 | 53.2 | 44.8 | 46.9 | |
| ARCANESize=32B, Mode=Vanilla2026.06 | 53.6 | 53.2 | 44.9 | 45 | |
| Qwen3Size=8B, Mode=RAG2026.06 | 53 | 51 | 43.2 | 44.6 | |
| Qwen3Size=32B, Mode=TimeChara2026.06 | 51.3 | 50.4 | 41.3 | 45.4 | |
| Qwen3Size=8B, Mode=LifeChoice2026.06 | 50.9 | 48.9 | 40.8 | 42.4 | |
| Qwen3Size=8B, Mode=Arc2026.06 | 50.1 | 48.5 | 39.2 | 42.1 | |
| Qwen3Size=32B, Mode=Vanilla2026.06 | 49.7 | 48.6 | 39.4 | 43.4 | |
| ARCANESize=8B, Mode=TimeChara2026.06 | 49.3 | 48.2 | 40.2 | 43 | |
| ARCANESize=8B, Mode=Vanilla2026.06 | 47.8 | 46.9 | 38.3 | 41.9 | |
| Qwen3Size=8B, Mode=Summary2026.06 | 47.2 | 45.4 | 36.5 | 41.7 | |
| Qwen3Size=8B, Mode=TimeChara2026.06 | 43.1 | 43.1 | 33.3 | 36.5 | |
| Qwen3Size=8B, Mode=Vanilla2026.06 | 43 | 42.8 | 32.7 | 38.2 |