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

Language Understanding on CEval

83.56Accuracy

Qwen3-30B-A3B

3.552824.323945.09565.8661Dec 25, 2025Jan 20, 2026Feb 16, 2026Mar 15, 2026Apr 11, 2026May 8, 2026Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.05
83.56
2026.05
81.46
2026.05
74.15
2026.05
70.47
2026.06
70.05
2026.06
69.54
2026.06
67.9
2026.05
67.13
2026.05
66.73
2026.05
66.11
2026.06
64.89
2026.06
64.56
2026.06
64.34
2025.12
63.03
2026.06
62
2026.06
61.28
2026.06
61.22
2026.05
61
2026.06
60.04
2026.06
60.03
2026.06
59.38
2026.05
59.3
2026.06
59.3
2025.12
59.16
2026.06
58.96
2026.06
58.79
2026.05
58.51
2026.06
58.43
2026.06
57.95
2026.06
57.66
2026.06
56.32
2026.06
55.89
2026.05
55.26
2026.06
54.21
2026.05
54.12
2026.06
53.27
2026.06
53.19
2026.02
50.81
2026.06
49.89
2026.02
49.06
2025.12
44.81
2026.05
42.6
2025.12
41.74
2026.05
40.14
2026.05
39.17
2026.05
37.81
2026.05
37.6
2026.05
36.42
2026.05
35.67
2025.12
34.83
2026.02
33.81
2026.02
33.75
2026.05
33.72
2026.02
33.45
2026.02
33.27
2026.02
32.88
2026.02
32.87
2026.02
32.8
2026.05
32.78
2026.02
32.53
2026.05
32.49
2026.02
32.08
2026.05
32.08
2026.02
29.38
2026.05
24.5
2026.05
23.8
2026.05
6.63