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

General Capability on MMLU

79.6MMLU Accuracy

Original Weight Model

44.2453.4262.671.78Apr 2, 2025Jun 8, 2025Aug 15, 2025Oct 21, 2025Dec 28, 2025Mar 5, 2026May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2025.04
79.6
2025.04
78.89
2025.10
76.1
2025.10
76.1
2025.10
74.7
2025.10
74.4
2025.10
74
2025.10
73.3
2025.10
72.8
2025.10
72.8
2025.10
72.3
2025.10
71.2
2025.10
70.9
2025.10
70.5
2025.10
70.2
2025.10
69.8
2025.10
68.1
2025.10
68.1
2025.10
68
2025.10
68
2025.10
67.7
2025.10
67.3
2025.10
66.7
2025.04
66.44
2025.04
66.29
2025.04
66.16
2025.04
66.13
2025.04
66.09
2025.04
66.01
2025.10
66
2025.04
65.89
2025.04
65.77
2025.04
65.77
2025.04
65.31
2025.04
65.08
2025.04
64.41
2025.10
63.4
2025.10
62.4
2025.10
62.3
2025.10
62.3
2026.05
62.22
2025.10
62.2
2025.10
62
2025.04
60.18
2025.04
60.08
2025.04
60.08
2025.04
59.93
2025.04
59.61
2025.04
59.48
2025.04
59.44
2025.04
59.26
2025.04
59.24
2025.04
59.21
2025.04
59.12
2025.04
58.57
2025.04
58.46
2025.04
58.23
2025.04
58.14
2025.04
57.98
2025.10
57.9
2025.10
57.9
2025.10
57.4
2025.10
57.2
2025.10
56.8
2025.10
56.8
2025.04
50.02
2025.10
49.1
2025.10
47.6
2025.10
47.6
2025.10
47.5
2025.10
47
2025.10
46
2025.10
46
2025.10
45.6