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

Multi-subject Knowledge Reasoning on MMLU-Pro

73.44Pass@1

FP16

49.57255.768561.96568.1615May 9, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.05
73.44
2026.05
72.67
2026.05
72.65
2026.05
72.59
2026.05
72.43
2026.05
72.33
2026.05
71.52
2026.05
70.23
2026.05
69.41
2026.05
68.73
2026.05
68.41
2026.05
68.28
2026.05
68.03
2026.05
68
2026.05
58.84
2026.05
57.31
2026.05
56.78
2026.05
56.77
2026.05
56.69
2026.05
56.3
2026.05
55.96
2026.05
55.58
2026.05
55.44
2026.05
54.98
2026.05
54.58
2026.05
54.44
2026.05
54.21
2026.05
50.49