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

Knowledge Evaluation on MMLU

82.03MMLU Accuracy

W16A16-Direct

48.937257.528666.1274.7114Feb 13, 2026Feb 23, 2026Mar 5, 2026Mar 15, 2026Mar 25, 2026Apr 4, 2026Apr 14, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
82.030
2026.04
81.98-0.06
2026.04
79.91-2.12
2026.04
79.78-2.25
2026.04
79.28-2.75
2026.04
78.93-3.1
2026.04
78.23-3.8
2026.04
74.970
2026.04
74.61-0.36
2026.02
74.26-
2026.02
73.65-
2026.02
73.41-
2026.02
73.31-
2026.04
72.64-2.33
2026.04
72.4-2.56
2026.04
72.26-2.71
2026.04
70.31-4.66
2026.04
68.24-6.73
2026.02
65.52-
2026.02
64.64-
2026.02
64.48-
2026.02
64.37-
2026.02
53.03-
2026.02
51.77-
2026.02
50.69-
2026.02
50.21-