Share your thoughts, 1 month free Claude Pro on usSee more

Medical Question Answering on MedMCQA (Generation Metrics)

10.82BLEU Score

Llama-3.1-8B-Instruct

Updated 1mo ago

Evaluation Results

Method
Llama-3.1-8B-Instruct 2026.03	10.82	36.55	25.97	88.49
glm4-9b-chat 2026.03	9.82	36.02	25.5	88.3
Mistral-7B-Instruct-v0.1 2026.03	9.71	36.11	25.2	88.37
Mistral-7B-Instruct-v0.3 2026.03	9.58	35.28	25.08	88.3
Llama-3.1-8B-Instruct 2026.03	8.81	30.54	21.36	87.31
Llama-3.1-8B-Instruct 2026.03	8.49	29.96	20.84	87.2
glm4-9b-chat 2026.03	8.48	29.92	21.67	87.13
Phi-3.5-mini-instruct 2026.03	8.3	29.18	20.05	87.07
Mistral-7B-Instruct-v0.3 2026.03	8.24	28.94	20.86	87.07
Llama-3.1-8B-Instruct 2026.03	8.21	30.17	21.11	87.14
Phi-3.5-mini-instruct 2026.03	8.05	29.69	19.95	87.21
Llama-2-7b-chat-hf 2026.03	7.99	29.78	20.14	87.15
Phi-3.5-mini-instruct 2026.03	7.61	27.85	18.59	86.81
Mistral-7B-Instruct-v0.1 2026.03	7.43	28.85	20.29	86.92
Phi-3.5-mini-instruct 2026.03	7.42	27.7	18.57	86.74
Mistral-7B-Instruct-v0.1 2026.03	7.06	27.86	19.16	86.7
Mistral-7B-Instruct-v0.1 2026.03	6.97	27.33	18.82	86.65
glm4-9b-chat 2026.03	6.9	28.27	19.67	86.8
Mistral-7B-Instruct-v0.3 2026.03	6.88	27.55	19.29	86.76
Qwen2.5-7B-Instruct 2026.03	6.8	24.07	16.58	85.95
Mistral-7B-Instruct-v0.3 2026.03	6.71	27.02	18.79	86.63
Phi-3.5-mini-instruct 2026.03	6.63	25.4	16.9	86.3
Qwen2.5-7B-Instruct 2026.03	6.63	27.16	18.79	86.54
glm4-9b-chat 2026.03	6.61	28.44	19.87	86.73
Qwen2.5-7B-Instruct 2026.03	6.3	24.13	16.8	85.93
Qwen2.5-7B-Instruct 2026.03	5.92	24.65	17.73	86.14
Llama-2-7b-chat-hf 2026.03	5.91	23.25	15.54	85.79
Llama-3.1-8B-Instruct 2026.03	5.72	23.29	16.18	85.71
Llama-2-7b-chat-hf 2026.03	5.56	23.77	16.06	85.92
Qwen1.5-7B-Chat 2026.03	5.46	23.72	15.78	85.84
Llama-2-7b-chat-hf 2026.03	5.37	23.76	15.8	85.9
Qwen1.5-7B-Chat 2026.03	4.85	21.55	14.84	85.4
Qwen2.5-7B-Instruct 2026.03	4.67	20.3	13.85	85.06
Qwen1.5-7B-Chat 2026.03	4.6	19.65	13.12	85
Mistral-7B-Instruct-v0.3 2026.03	4.32	19.17	13.11	85.03
glm4-9b-chat 2026.03	4.17	20.62	13.86	85.17
Qwen1.5-7B-Chat 2026.03	4.13	19.34	12.86	84.9
Mistral-7B-Instruct-v0.1 2026.03	3.35	18.09	12.03	84.74
Llama-2-7b-chat-hf 2026.03	3.34	16.13	10.45	84.35
Qwen1.5-7B-Chat 2026.03	3.11	15.09	9.95	84.1
DeepSeek-R1-Distill-Qwen-7B 2026.03	2.85	14.53	9.2	84
DeepSeek-R1-Distill-Qwen-7B 2026.03	2.62	13.55	8.91	84
DeepSeek-R1-Distill-Qwen-7B 2026.03	2.44	12.55	7.92	83.7
DeepSeek-R1-Distill-Qwen-7B 2026.03	1.91	11.99	7.54	83.62
DeepSeek-R1-Distill-Qwen-7B 2026.03	1.85	10.41	6.5	83.19
Mistral-7B-Instruct-v0.1 2026.03	0.12	3.17	6.29	79.6
Mistral-7B-Instruct-v0.3 2026.03	0.11	2.1	5.19	79.22
Llama-3.1-8B-Instruct 2026.03	0.1	2.34	5.01	78.58
Phi-3.5-mini-instruct 2026.03	0.09	1.59	4.1	78.44
Qwen2.5-7B-Instruct 2026.03	0.09	1.8	4.34	78.35
glm4-9b-chat 2026.03	0.08	1.71	4.21	78.48
Qwen1.5-7B-Chat 2026.03	0.08	1.64	4.03	78.63
Llama-2-7b-chat-hf 2026.03	0.06	0.99	2.54	77.33
DeepSeek-R1-Distill-Qwen-7B 2026.03	0.02	0.52	1.31	76.29