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

Multi-task Language Understanding on MMLU (Accuracy and vNMSE)

73.04Accuracy

BF16

71.53271.923572.31572.7065Feb 9, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
73.040
2026.02
73.040.0007
2026.02
72.860.002
2026.02
72.460.0201
2026.02
71.590.1706