List your company

Language Modeling on OOD

1.285LossMuon
1.279081.319041.3591.39896Apr 10, 2026

Evaluation results

MethodLinks
MuonOptimizer=Muon, Model Size=3B2026.041.285
Adam+NexusOptimizer=Adam+Nexus, Model Size=3B2026.041.29
NexusModel Size=3B, Optimizer=Nexus2026.041.29
AdamWOptimizer=AdamW, Model Size=3B2026.041.302
AdamWModel Size=3B, Optimizer=AdamW2026.041.302
NexusModel Size=1B, Optimizer=Nexus2026.041.428
AdamWModel Size=1B, Optimizer=AdamW2026.041.433