Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language Modeling on GPT Pre-training (val)
Loading...
19.98
Validation Perplexity
MUON+
19.5916
22.2133
24.835
27.4567
Feb 25, 2026
Validation Perplexity
Updated 19d ago
Evaluation Results
Method
Method
Links
Validation Perplexity
MUON+
Model Scale=GPT-Base
2026.02
19.98
NorMuon
Model Scale=GPT-Base
2026.02
21.31
Turbo-Muon
Model Scale=GPT-Base
2026.02
21.91
AdaMuon
Model Scale=GPT-Base
2026.02
22.38
MUON+
Model Scale=GPT-Small
2026.02
27.64
NorMuon
Model Scale=GPT-Small
2026.02
28.44
AdaMuon
Model Scale=GPT-Small
2026.02
29.27
Turbo-Muon
Model Scale=GPT-Small
2026.02
29.69
Feedback
Search any
task
Search any
task