Our new X account is live! Follow @wizwand_team for updates

Role-playing on CharacterBench 1.0 (test)

4.444MC

Qwen2.5-7B-Instruct + Character-R1

Updated 4d ago

Evaluation Results

Method
Qwen2.5-7B-Instruct + Character-R1 2026.01	4.444	2.688	4.017	4.381	4.544	3.317	3.956	3.463	3.072	4.925	4.861	3.03	3.115	3.832
Llama-3.2-3B-Instruct + Character-R1 2026.01	4.275	2.144	3.983	4.556	4.95	3.35	4.038	3.463	3.134	4.85	4.748	2.94	2.97	3.8
Qwen2.5-7B-RoleMRC 2026.01	4.175	2.231	3.775	4.3	4.713	3.225	3.688	3.188	2.927	4.75	4.735	3	3.105	3.678
Llama-3.1-8B-RoleMRC 2026.01	4.169	2.163	3.717	4.231	4.588	3.225	3.606	3.125	2.864	4.8	4.748	3.04	3.06	3.641
Crab 2026.01	4.131	2.175	3.6	4.25	4.65	3.167	3.681	3.331	2.965	4.85	4.684	3.05	3.11	3.665
Haruhi-Zero-7B 2026.01	4.088	2.15	3.583	4.256	4.525	3.192	3.613	3.456	2.965	4.85	4.773	2.99	3.065	3.654
Qwen2.5-72B-Instruct 2026.01	4.06	2.558	4.102	3.222	4.531	3.439	3.917	3.398	2.962	4.897	4.815	3.512	3.418	3.756
Qwen2.5-7B-Instruct + RAR 2026.01	4.059	2.585	3.934	4.272	4.314	3.977	4.09	3.145	2.809	4.825	4.808	2.814	2.946	3.737
CoSER-Llama-3.1-8B 2026.01	4.056	2.113	3.958	4.213	4.506	3.25	3.7	3.156	2.883	4.95	4.874	3.045	2.97	3.667
Qwen2.5-7B-Instruct + SFT 2026.01	4.02	2.165	3.873	4.105	4.24	3.207	3.545	3.17	2.771	4.72	4.788	3	3.056	3.589
Humanish-Roleplay 2026.01	4.013	2.35	4	4.481	4.694	3.267	3.975	3.444	3.053	4.75	4.823	3.01	3.145	3.77
Qwen2.5-7B-Instruct + Neeko 2026.01	3.997	2.192	3.775	3.288	4.257	3.084	3.212	3.229	2.693	4.55	4.703	2.799	2.942	3.44
Claude-4-opus 2026.01	3.997	2.435	4.398	4.362	4.508	3.751	3.877	3.587	3.176	4.936	4.735	3.151	3.351	3.866
Gemini-3-pro 2026.01	3.946	2.905	4.099	4.133	4.167	3.529	3.671	3.319	2.973	4.82	4.633	3.613	3.423	3.787
Llama-3-70B-Instruct 2026.01	3.94	2.59	3.95	3.96	4.39	3.35	3.33	3.06	2.89	4.71	4.74	3.4	3.51	3.678
Qwen2.5-7B-Instruct 2026.01	3.938	2.356	3.8	4.275	4.631	3.233	3.856	3.238	3.097	4.918	4.861	3.12	3.12	3.726
Llama-3.2-3B-Instruct + RAR 2026.01	3.848	2.01	3.978	4.373	4.511	4.112	3.918	3.24	2.91	4.785	4.743	2.718	2.964	3.701
Llama-3.2-3B-Instruct + Neeko 2026.01	3.796	1.855	3.816	3.428	4.513	3.192	3.416	3.319	2.747	4.561	4.582	2.665	2.972	3.451
GPT-4o 2026.01	3.793	2.647	3.978	4.027	4.484	3.414	3.723	3.046	2.974	4.763	4.771	3.261	3.51	3.722
CharacterGLM 2026.01	3.76	2.18	3.97	3.8	4.03	2.89	3.26	2.94	2.64	4.53	4.51	3.16	3.32	3.461
Baichuan-NPC 2026.01	3.672	2.134	4.132	4.216	4.254	3.366	4.022	3.001	3.179	4.83	4.897	2.975	3.297	3.69
Llama-3.2-3B-Instruct + SFT 2026.01	3.606	2.063	3.892	4.169	4.55	3.5	3.894	3.219	3.003	4.888	4.823	3.35	3.31	3.713
Llama-3.2-3B-Instruct 2026.01	3.588	2.088	3.892	4.3	4.606	3.3	4.025	3.306	3.053	4.84	4.649	2.89	3.32	3.681
GLM-4 2026.01	3.524	2.373	3.701	4.103	4.38	3.487	3.728	3.13	2.987	4.826	4.876	3.208	3.5	3.679
GPT-3.5-turbo 2026.01	3.49	2.451	3.692	4.155	4.345	3.56	3.635	3.09	2.838	4.735	4.761	3.619	3.758	3.702
CharacterGLM-6B 2026.01	3.245	2.1	3.543	3.41	3.365	3.1	3.07	2.61	2.5	4.48	4.8	2.84	2.7	3.213
Peach-9B-8k-Roleplay 2026.01	2.931	2.344	3.717	3.388	3.8	3.292	3.3	2.863	2.802	4.838	4.798	2.475	2.435	3.306
Peach-2.0-9B-8k-Roleplay 2026.01	2.819	2.4	3.425	4.263	4.4	3.867	4.088	3.35	3.028	4.475	4.47	3.485	3.68	3.673