Ultrasound Report Generation on Thyroid Ultrasound Examinations benchmark leaderboard · SOTA2 Research

Ultrasound Report Generation on Thyroid Ultrasound Examinations

79.6BLEU-1LUMI

Evaluation Results

Method	Links
LUMI2026.07		79.6	73.8	68.2	63	75.7	78	93.5	66.7
Qwen3.5-27B2026.07		36.2	22.6	14.6	9.4	26.8	25.4	73.7	33.5
InternVL3.5-30B-A3B-Instruct2026.07		33.1	21	13.9	9.5	25.6	23.7	72.8	28.5
InternVL3.5-8B-Instruct2026.07		32.3	20.7	13.6	8.9	26.1	24	74.4	33.4
InternVL3.5-38B-Instruct2026.07		31.1	18.9	12.5	8.4	23.9	21.3	70.9	27.8
Qwen3.5-9B2026.07		30.7	18.1	11.6	7.5	23.4	23.2	72	32.3
InternVL3.5-14B-Instruct2026.07		29.5	18.7	12.6	8.6	24.5	21.5	71.9	22.8
Lingshu-32B2026.07		29.4	18.3	11.9	7.8	26.8	22.9	73.3	22.3
Qwen3.5-35B-A3B2026.07		28.4	17.2	10.6	6.6	24.3	26	72.6	33.8
InternVL3.5-1B-Instruct2026.07		25.7	16	10.7	7.5	25.2	21.9	72.4	28.7
HuatuoGPT-Vision-34B2026.07		25.3	16.4	11.3	8.1	26.6	22.3	73.6	31.7
Lingshu-7B2026.07		25.2	15.5	10.5	7	23.9	18.4	70.8	33.9
Qwen3.5-0.8B2026.07		24.7	13.4	8.5	5.4	19.9	16.6	69.5	20.2
Qwen3.5-2B2026.07		23.8	13.4	8.5	5.5	20.5	20.5	70.1	21.9
Qwen3.5-4B2026.07		22.9	12.5	7.5	4.4	20.3	22	70.6	17.8
InternVL3.5-2B-Instruct2026.07		22.7	14	9.2	6.1	25	22.1	71.7	30.6
EchoVLM2026.07		22.3	12.9	7.8	4.3	29.3	21.8	73.1	28.6
InternVL3.5-4B-Instruct2026.07		21	13.4	8.8	6	26	21.9	72.4	29.8
medgemma-4b-it2026.07		13.1	5.9	3.2	1.4	13.5	14.3	63.2	14.9
HuatuoGPT-Vision-7B2026.07		8.4	5.4	3.9	2.7	23.7	14.2	70.3	45.7
Kimi-VL-A3B-Instruct-16B2026.07		5.7	3.5	2.3	1.6	26.4	18.9	71.6	23.5