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

Multi-subject Knowledge on MMLU

79.7Accuracy

Baseline

43.8253.13562.4571.765Dec 10, 2025Jan 4, 2026Jan 30, 2026Feb 24, 2026Mar 22, 2026Apr 16, 2026May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
79.7
2026.05
79.4
2026.05
77.7
2025.12
74.26
2025.12
74.23
2025.12
74.22
2025.12
74.2
2025.12
74.14
2025.12
69.38
2025.12
69.21
2025.12
69.21
2025.12
69.16
2025.12
69.14
2026.05
48
2026.05
45.4
2026.05
45.2