List your company

Knowledge on MMLU (test)

32.92Accuracy (Knowledge, test)MoE w/ optimal AR
24.23626.490528.74530.9995Jun 13, 2025

Evaluation results

MethodLinks
MoE w/ optimal ARModel Type=MoE, Activation rate=20.07, Compute=2.86e21, Data reuse=-2025.0632.92
Dense baselineModel Type=Dense, Compute=5.45e212025.0631.26
MoE w/ optimal ARModel Type=MoE, Activation rate=20.07, Compute=2.86e21, Data reuse=strict2025.0624.57