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

Multi-task Language Understanding on MMLU (MMLU, Avg.5)

80.2MMLU Score

QWEN-3-30B MOE

2.210422.457742.70562.9523Nov 14, 2025Dec 15, 2025Jan 15, 2026Feb 15, 2026Mar 18, 2026Apr 18, 2026May 20, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2025.11
80.2-
2025.11
80-
2025.11
75.9-
2025.11
74-
2026.05
72.4366.42
2025.11
69.7-
2026.05
68.3263.72
2025.11
63.1-
2026.04
60.8746.1
2026.05
59.8-
2026.05
57-
2025.11
56.8-
2026.05
55.9746.93
2026.05
55.3-
2026.05
54.8-
2026.05
54.38-
2026.05
54.25-
2026.05
54.19-
2026.05
54.1-
2026.05
54.1-
2026.04
53.433.45
2026.05
52.98-
2026.05
52.88-
2026.05
52.79-
2026.05
52.33-
2025.11
50.5-
2026.05
50.1-
2026.05
49.2-
2026.04
49.1137.48
2026.04
48.7237.54
2026.05
48.6-
2026.05
48.5-
2025.11
48-
2026.05
47.9-
2026.05
47.43-
2026.05
47.23-
2026.05
47.1-
2026.05
47.07-
2026.05
46.3240.48
2026.05
46-
2026.05
45.8640.15
2026.05
45.74-
2026.05
45.1-
2026.05
45.1-
2026.05
44.7-
2026.05
44.6738.85
2026.05
44.15-
2026.05
43.9838.49
2026.05
43.9339.18
2026.05
43.8338.4
2026.05
43.538.66
2026.04
43.0824.85
2026.05
42.5635.03
2026.04
42.3324.54
2026.05
42.2-
2026.05
41.4338.31
2026.05
40.536.63
2026.05
40.3436.77
2026.05
40.3-
2026.04
40.1930.6
2026.04
39.0128.8
2026.04
38.4426.94
2026.04
34.7522.56
2026.05
32.0533.96
2026.04
31.62-
2026.04
31.18-
2026.04
30.78-
2026.04
30.1716.97
2026.04
28.315.1
2026.04
28.0114.85
2026.04
26.9416.96
2026.04
26.41-
2026.04
26.36-
2026.04
26.22-
2026.04
2639.4
2026.04
2642
2026.04
24.4611.77
2026.04
2442.9
2026.04
2447.4
2026.04
2344.3
2026.04
2244.1
2026.04
20.159.69
2026.04
15.2212.1
2026.04
14.786.07
2026.04
8.33.88
2026.04
5.211.9