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

Medical Visual Question Answering on MicroVQA

67.1Overall Accuracy

Qwen3-VL-8B-Instruct

Updated 21d ago

Evaluation Results

Method
Qwen3-VL-8B-Instruct 2026.05	67.1	-	-	-	67.9	68.4	63.5
GPT-o4-mini 2026.01	65	68.75	67.78	54.35	-	-	-
Qwen3-VL-4B-Instruct 2026.05	64.3	-	-	-	62.9	66.6	63
InternVL3.5-8B-Instruct 2026.05	63.4	-	-	-	64	63.5	62.2
Gemini-2.5-pro 2026.01	60.9	61.2	62.6	57.4	-	-	-
GPT-5 2026.01	60.8	62	61.2	58.3	-	-	-
GPT-5 2026.05	59.4	-	-	-	58.9	53.9	63.3
Qwen3-VL-2B-Instruct 2026.05	59.4	-	-	-	60	57.4	61.7
o3 2026.05	59.3	-	-	-	60.5	53.5	61.5
Qwen3-VL-8B-Instruct 2026.05	58.9	-	-	-	58.3	60.5	57.4
GPT-5-mini 2026.01	57.1	58.7	57.61	53.51	-	-	-
Claude-4.5-sonnet 2026.01	56.33	55.6	57.91	54.83	-	-	-
InternVL3.5-2B-Instruct 2026.05	56.3	-	-	-	53.3	55.4	63.5
o4-mini 2026.05	55.6	-	-	-	56.1	50.4	57.9
InternVL3.5-8B-Instruct 2026.05	54.8	-	-	-	55	55.6	53
Qwen3-VL-4B-Instruct 2026.05	54.7	-	-	-	55.2	55.1	53
Claude Sonnet 4.5 2026.05	54.4	-	-	-	56.4	49.6	55.1
Claude-4.5-haiku 2026.01	53.14	52.81	51.21	57	-	-	-
o1 2026.05	52.8	-	-	-	50.2	53	55.4
Qwen3-VL-2B-Instruct 2026.05	52.8	-	-	-	51.9	51.8	56.1
InternVL3.5-2B-Instruct 2026.05	51.3	-	-	-	49.5	49.2	58.3
InternVL3.5-4B-Instruct 2026.05	49.8	-	-	-	48.1	51	50.9
Qwen3-VL-8B-Instruct 2026.05	48.8	-	-	-	48.6	49.2	48.3
Gemini-2.5-flash 2026.01	47	50	44.5	46.5	-	-	-
Qwen3-VL-4B-Instruct 2026.05	46.4	-	-	-	46.4	47.4	44.8
Qwen3-VL-2B-Instruct 2026.05	45.9	-	-	-	42.4	44.9	53.9
InternVL3.5-4B-Instruct 2026.05	45.6	-	-	-	43.1	45.4	50.4
InternVL3.5-2B-Instruct 2026.05	44.5	-	-	-	42.9	43.4	49.6
InternVL3.5-4B-Instruct 2026.05	44.4	-	-	-	43.8	45.9	43
GRPO w/o Tools 2026.01	43	40.63	44.44	43.48	-	-	-
MEDVISTA-R1 2026.01	43	42.2	37.8	54.41	-	-	-
Cold-start w/o Tools 2026.01	42.5	43.75	43.33	39.1	-	-	-
InternVL3.5-8B-Instruct 2026.05	41.8	-	-	-	38.1	42.6	47.4
Direct GRPO w/o cold-start 2026.01	41	43.75	38.89	41.3	-	-	-
MEDVISTA-R1 2026.01	39.5	37.5	37.78	45.65	-	-	-
Cold-start w/o Reasoning 2026.01	39	39.1	37.8	41.3	-	-	-
Direct GRPO w/o cold-start 2026.01	38.5	35.9	36.67	43.48	-	-	-
GRPO w/o Tools 2026.01	37	34.38	37.78	39.13	-	-	-
Cold-start w/o Reasoning 2026.01	37	35.9	38.9	34.8	-	-	-
MEDVISTAGYM (Qwen3vl-8B) 2026.01	36.28	40.62	35.61	32.63	-	-	-
Qwen3vl-8B 2026.01	35.9	38.8	31.9	38.3	-	-	-
Cold-start w/o Tools 2026.01	35.5	39.1	41.11	34.78	-	-	-
MEDVISTAGYM (InternVL3-8B) 2026.01	35	39.06	32.2	34.8	-	-	-
Qwen2.5vl-7B 2026.01	33	34.38	31.11	34.78	-	-	-
InternVL3-8B 2026.01	31.5	31.9	26.7	39.6	-	-	-
MEDVISTAGYM (InternVL3-2B) 2026.01	31	34.4	25.6	37	-	-	-
Internvl3-2B 2026.01	30.48	30	28.57	34.86	-	-	-
LLaVA-Med-7B 2026.01	16.51	12.2	17.6	21.7	-	-	-