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

Multitask Language Understanding on MMMLU (non-EU languages, test)

77.4Accuracy

Qwen-3-30B-A3B

45.1653.5361.970.27Feb 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
77.4
2026.02
77
2026.02
75.5
2026.02
72.5
2026.02
72.4
2026.02
72.1
2026.02
66.6
2026.02
63
2026.02
61.1
2026.02
57.7
2026.02
57.6
2026.02
56.6
2026.02
51.3
2026.02
50.9
2026.02
50.2
2026.02
46.4