Visual Question Answering on FetUS-VQA benchmark leaderboard · SOTA2 Research

Visual Question Answering on FetUS-VQA

99.42Task 1 PerformanceFetUSAgents

Evaluation Results

Method	Links
FetUSAgents2026.05		99.42	70.46	93.32	89.58	93.96	92.77	91.02	91.02	100	50.51	87.21
Qwen3-MAX2026.05		42.97	37.25	33.68	24.16	40.7	10.94	20.2	27.89	26.51	25.9	29.02
Doubao-Seed-2.0-Pro2026.05		40.78	52.71	62.02	46.38	81.97	69.06	72.85	29.01	33.21	29.68	51.77
Qwen3-VL-30B-A3B2026.05		40.73	30.91	39.28	29.8	61.94	35.9	40.26	42.71	29.75	27.89	37.92
Hulu-Med-32B2026.05		32.48	55.86	56.01	27.62	70.38	54.07	43.64	59.28	20.42	29.49	44.93
GPT-5.42026.05		31.65	46.88	55	38.66	78.05	71.76	76.85	70.33	32.61	10.21	51.2
Gemini-3-Flash-Thinking2026.05		31.48	35.92	70.12	68.93	91.84	88.03	57.79	45.89	41.49	19.39	55.09
Hulu-Med-30A32026.05		29.44	43.25	58.87	19.95	58.3	63.45	38.74	42.71	15.56	15.57	38.58
GPT-5.12026.05		28.8	52.05	54.84	34.01	74.19	64.2	74.98	32.07	34.34	25.6	47.51
DeepSeek-V3.22026.05		24.49	44.65	33.37	19.76	47.82	16.3	42.71	28.26	19.47	21.48	29.83
Llava-Med-V1.5-Mistral-7B2026.05		17.96	37.25	40.71	20.41	33.78	17.47	20.28	42.71	12.53	11.82	25.49
HuatuoGPT-34B2026.05		17.89	35.8	47.81	29.18	64.31	27.53	35.91	56.79	17.71	20.43	35.34
MedGemma-27B2026.05		17.62	51.94	56.91	26.32	56.9	30.54	42.71	42.71	14.91	22.84	36.34