Character Roleplay on ARCANE validated slice (5 novels) (In-world)
66.2APFARCANE
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ARCANESize=32B, Mode=Arc2026.06 | 66.2 | 64.2 | 58.8 | 58.5 | |
| DeepSeek-V4Size=Pro, Mode=Arc2026.06 | 64.2 | 63.6 | 57 | 58.2 | |
| DeepSeek-V4Size=Flash, Mode=Arc2026.06 | 61.8 | 61.9 | 53.8 | 55.3 | |
| ARCANESize=8B, Mode=Arc2026.06 | 60.9 | 59.3 | 52.3 | 54.6 | |
| DeepSeek-V4Size=Pro, Mode=Summary2026.06 | 59.7 | 59.5 | 51.8 | 51.2 | |
| DeepSeek-V4Size=Pro, Mode=LifeChoice2026.06 | 59.1 | 58.9 | 51.1 | 50.9 | |
| DeepSeek-V4Size=Flash, Mode=Summary2026.06 | 57.8 | 58.1 | 49.4 | 49.4 | |
| DeepSeek-V4Size=Pro, Mode=RAG2026.06 | 57.6 | 57.6 | 49.2 | 50.5 | |
| DeepSeek-V4Size=Flash, Mode=LifeChoice2026.06 | 57.4 | 57.8 | 48.6 | 49.5 | |
| DeepSeek-V4Size=Flash, Mode=Vanilla2026.06 | 56.7 | 56.9 | 48.8 | 45.4 | |
| DeepSeek-V4Size=Flash, Mode=RAG2026.06 | 56.6 | 57.1 | 48 | 47.8 | |
| DeepSeek-V4Size=Flash, Mode=TimeChara2026.06 | 56.2 | 56.6 | 48.4 | 46.8 | |
| ARCANESize=32B, Mode=TimeChara2026.06 | 55.6 | 55 | 47.5 | 41.8 | |
| ARCANESize=32B, Mode=Vanilla2026.06 | 55.1 | 55.1 | 46.7 | 41.7 | |
| DeepSeek-V4Size=Pro, Mode=TimeChara2026.06 | 54.9 | 55.4 | 46.9 | 46.7 | |
| ARCANESize=32B, Mode=LifeChoice2026.06 | 54.9 | 54.2 | 45.8 | 44.3 | |
| DeepSeek-V4Size=Pro, Mode=Vanilla2026.06 | 54.2 | 54.9 | 46.1 | 45.5 | |
| Qwen3Size=32B, Mode=Arc2026.06 | 54.1 | 53.7 | 44.3 | 41 | |
| ARCANESize=32B, Mode=Summary2026.06 | 53.2 | 52.6 | 44 | 44.3 | |
| ARCANESize=32B, Mode=RAG2026.06 | 52.9 | 52.9 | 43.8 | 41.9 | |
| Qwen3Size=32B, Mode=LifeChoice2026.06 | 50.6 | 50.5 | 40.9 | 40.3 | |
| Qwen3Size=32B, Mode=RAG2026.06 | 49.6 | 49.9 | 40 | 38.6 | |
| Qwen3Size=32B, Mode=Summary2026.06 | 49.5 | 49.3 | 39.2 | 36 | |
| ARCANESize=8B, Mode=Summary2026.06 | 49.4 | 48.8 | 40 | 40.3 | |
| ARCANESize=8B, Mode=LifeChoice2026.06 | 49.1 | 48.6 | 40.1 | 39.3 | |
| Qwen3Size=32B, Mode=TimeChara2026.06 | 48.9 | 49.1 | 39.5 | 36.4 | |
| ARCANESize=8B, Mode=RAG2026.06 | 48.8 | 48.8 | 40 | 39.6 | |
| Qwen3Size=32B, Mode=Vanilla2026.06 | 48.6 | 48.6 | 38.8 | 35.8 | |
| ARCANESize=8B, Mode=Vanilla2026.06 | 48.2 | 47.5 | 39 | 37.4 | |
| ARCANESize=8B, Mode=TimeChara2026.06 | 47.8 | 47.3 | 38.8 | 38.9 | |
| Qwen3Size=8B, Mode=Arc2026.06 | 47.5 | 47.3 | 37.3 | 36.6 | |
| Qwen3Size=8B, Mode=RAG2026.06 | 43 | 43.1 | 33.1 | 31.6 | |
| Qwen3Size=8B, Mode=LifeChoice2026.06 | 42.6 | 42.3 | 32.1 | 31.4 | |
| Qwen3Size=8B, Mode=Summary2026.06 | 42.1 | 41.9 | 32.1 | 31.6 | |
| Qwen3Size=8B, Mode=TimeChara2026.06 | 41.6 | 42.2 | 32 | 30.6 | |
| Qwen3Size=8B, Mode=Vanilla2026.06 | 40.5 | 41 | 30.8 | 30.6 |