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

Multimodal Speech Understanding on MMSU All (test)

75.2Accuracy

Step-Audio-R1

9.295226.405143.51560.6249May 26, 2026
Updated 6d ago

Evaluation Results

MethodLinks
75.2
2026.05
71.88
2026.05
71.28
64.16
2026.05
62.59
2026.05
59.92
57.16
2026.05
55.43
2026.05
53.3
2026.05
51.43
50.82
2026.05
50.74
2026.05
50.62
2026.05
50.36
2026.05
50.04
2026.05
49.98
2026.05
47.71
2026.05
47.1
2026.05
41.35
2026.05
11.83