List your company

GPU Utilisation on GLUE fine-tuning base

96GPU UtilisationMoE-MPMC
-2.48823.08148.6574.219May 12, 2026

Evaluation results

MethodLinks
MoE-MPMCModel Architecture=switch-base, Number of Experts=82026.0596
SiDA-MoEModel Architecture=switch-base, Number of Experts=82026.0589
MoE-MPMCModel Architecture=switch-base, Number of Experts=642026.0589
MoE-MPMCModel Architecture=switch-base, Number of Experts=1282026.0582
SiDA-MoEModel Architecture=switch-base, Number of Experts=642026.0579
SiDA-MoEModel Architecture=switch-base, Number of Experts=1282026.0563
SwitchTransformers (Baseline)Model Architecture=switch-base, Number of Experts=82026.057.5
SwitchTransformers (Baseline)Model Architecture=switch-base, Number of Experts=642026.053.4
SwitchTransformers (Baseline)Model Architecture=switch-base, Number of Experts=1282026.051.3