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

Multitask Language Understanding on MMLU CoT

72.8MMLU (CoT)

FP16

54.350459.140263.9368.7198Sep 27, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
72.8
2025.09
72.5
2025.09
72.4
2025.09
72.4
2025.09
71.8
2025.09
69.71
2025.09
69.4
2025.09
68.9
2025.09
68.9
2025.09
68.7
2025.09
68.69
2025.09
68.6
2025.09
68.59
2025.09
68.56
2025.09
68.2
2025.09
67.6
2025.09
67.2
2025.09
67
2025.09
66.8
2025.09
66.6
2025.09
65.48
2025.09
65
2025.09
64.5
2025.09
63.74
2025.09
62.6
2025.09
62.2
2025.09
61.22
2025.09
58.44
2025.09
55.06