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

General Knowledge Reasoning on MMLU (5-shot)

80.515-shot Accuracy

Base

58.597264.286169.97575.6639Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
80.5167.63
2026.07
79.65-
2026.07
78.6564.91
2026.07
78.32-
2026.07
78.31-
2026.07
78.17-
2026.07
78.13-
2026.07
78.160.22
2026.07
77.95-
2026.07
77.0858.1
2026.07
76.860.28
2026.07
76.7860.13
2026.07
75.9453.42
2026.07
69.63-
2026.07
69.55-
2026.07
69.54-
2026.07
68.93-
2026.07
68.72-
2026.07
68.69-
2026.07
62.11-
2026.07
61.69-
2026.07
61.27-
2026.07
60.77-
2026.07
60.6-
2026.07
59.44-