Benchmarks · creative
Japanese-RP-Bench v2 (tegnike fork)
Fork of Japanese-RP-Bench adding v2 metrics for role fidelity, conversation quality, persona stability, robustness, and recovery in Japanese roleplay. Maintained by tegnike. Score: Challenge RP Summary: macro-average of five criteria; leaderboard first sorts Major-free rate descending, then Major rate ascending.
15 models · 15 results · data as of 2026-10-29 · raw.githubusercontent.com ↗
Leaderboard
Grok 4.5 max
All results
15 rows| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Grok 4.5 | max | 95.8 |
| 2 | Grok 4.6 | max | 95.3 |
| 3 | Kimi K2.6 | max | 95.1 |
| 4 | Hy3 | xhigh | 94.6 |
| 5 | GLM 5.2 | max | 93.7 |
| 6 | MiniMax M3 | max | 93.6 |
| 7 | GPT-5.6 Luna | max | 90.0 |
| 8 | Kimi K3 | max | 89.6 |
| 9 | Qwen 3.7 Max | max | 88.2 |
| 10 | DeepSeek V4 Pro | max | 88.2 |
| 11 | Qwen 3.8 Max | max | 87.7 |
| 12 | DeepSeek V4 Flash | max | 86.6 |
| 13 | MiMo V2.5 Pro | xhigh | 85.7 |
| 14 | Qwen 3.7 Plus | high | 85.4 |
| 15 | MiniMax M2.7 | high | 80.5 |