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

Knowledge Reasoning on MMLU (Accuracy)

92.16MMLU Knowledge Reasoning Accuracy

SIGMA

29.91646.075562.23578.3945Oct 21, 2025Nov 25, 2025Dec 30, 2025Feb 3, 2026Mar 10, 2026Apr 14, 2026May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
92.16
2026.05
91.53
2026.05
90.85
2026.05
90.03
2026.02
89.85
2026.05
89.77
2026.05
89.41
2026.05
88.74
2026.05
87.46
2026.04
87.2
2026.04
86.8
2026.02
85.04
2026.02
84.65
2026.02
84.63
2026.02
84.4
2026.02
84.2
2026.02
84.1
2026.02
83.8
2026.04
83.5
2026.02
83.28
2026.02
83.25
2026.02
83.22
2026.02
83.2
2026.02
82.95
2026.04
82.8
2026.02
82.75
2026.02
82.4
2026.04
82.4
2026.02
81.85
2026.02
81.55
2026.02
80.47
2026.02
79.64
2026.04
78.6
2026.04
77.9
2026.04
74.5
2026.04
71
2026.04
70.1
2026.04
69.8
2026.04
68.8
2026.04
68.7
2026.04
66.6
2026.04
66.4
2025.10
61.91
2026.04
59.5
2026.04
58.5
2026.04
57.1
2026.04
56.7
2026.04
56.5
2026.04
55.8
2026.04
55.5
2026.04
55.2
2026.04
54.9
2026.04
54.8
2026.04
54.6
2026.04
53.9
2026.04
53.7
2025.10
52.52
2026.04
49.9
2026.04
49.1
2026.04
48.8
2026.04
48.5
2026.04
48.1
2026.04
48.1
2026.04
47
2026.04
46.5
2026.04
45.9
2026.04
45.5
2026.04
44.6
2026.04
44.2
2026.04
44
2026.04
43.7
2026.04
43.5
2025.10
32.31