Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Knowledge and Commonsense Reasoning on MMLU

89.1Accuracy

GPT-4o

23.621640.620857.6274.6192Sep 30, 2024Dec 17, 2024Mar 5, 2025May 22, 2025Aug 8, 2025Oct 25, 2025Jan 12, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2024.10
89.1
2024.10
86.4
2024.10
85.9
2024.10
78.9
2024.10
73.3
2024.10
69.4
2024.10
69.2
2026.01
60.6
2026.01
60.4
2024.09
58.25
2024.09
58.07
2024.09
57.72
2026.01
57.4
2024.09
57.37
2024.09
55.61
2026.01
48.6
2024.09
47.89
2024.09
42.81
2024.09
41.23
2024.09
39.3
2024.09
37.54
2024.09
28.77
2024.09
26.14