Clinical Diagnosis on S3 1.0 (test) benchmark leaderboard · SOTA2 Research

Clinical Diagnosis on S3 1.0 (test)

98PrecisionQwen3-VL-32B

Evaluation Results

Method	Links
Qwen3-VL-32B2026.05		98	80	88	85
Qwen2.5-VL-72B2026.05		98	75	85	82
seizure_omni_sft-7B2026.05		93	99	96	91
Qwen2.5-VL-32B2026.05		91	70	79	74
Qwen3-Omni-30B2026.05		91	98	94	89
seizure_omni_grpo-7B2026.05		91	98	94	90
Lingshu-32B2026.05		89	45	60	59
Qwen2.5-Omni-7B2026.05		88	100	93	88
Qwen3-Omni-30B2026.05		87	84	85	80
seizure_omni_grpo-7B2026.05		79	74	77	66
Qwen3-VL-8B2026.05		78	82	80	72
Qwen2.5-VL-7B2026.05		73	77	75	65
Qwen2.5-VL-32B2026.05		73	96	83	73
Qwen3-VL-32B2026.05		73	62	67	59
Lingshu-32B2026.05		73	100	84	74
InternVL3.5-8B2026.05		72	91	80	70
InternVL3.5-38B2026.05		72	96	82	72
Qwen2.5-VL-72B2026.05		72	91	80	69
Qwen2.5-Omni-7B2026.05		72	79	75	65
seizure_omni_grpo-7B2026.05		72	86	78	66
Qwen3-VL-32B2026.05		71	98	83	72
Qwen2.5-VL-72B2026.05		71	70	70	60
InternVL3.5-8B2026.05		70	98	81	69
Qwen2.5-VL-7B2026.05		70	55	62	54
Qwen3-Omni-30B2026.05		70	77	74	62
Lingshu-32B2026.05		70	93	80	68
Qwen2.5-VL-7B2026.05		69	96	81	68
InternVL3.5-38B2026.05		69	95	80	67
seizure_omni_sft-7B2026.05		69	94	79	66
Qwen3-VL-8B2026.05		68	100	81	68
Qwen2.5-Omni-7B2026.05		67	93	78	63
InternVL3.5-8B2026.05		66	77	71	57
seizure_omni_sft-7B2026.05		66	78	71	56
InternVL3.5-38B2026.05		65	46	54	46
Qwen2.5-VL-32B2026.05		64	75	69	54
Qwen3-VL-8B2026.05		62	59	61	48