Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Language Modeling on Wikipedia
Loading...
11.64
Perplexity
MIDUS-HML
11.5692
12.0471
12.525
13.0029
Dec 15, 2025
Perplexity
Updated 4d ago
Evaluation Results
Method
Method
Links
Perplexity
MIDUS-HML
Training Stage=CPT-1B,...
2025.12
11.64
OpT-DeUS
Training Stage=CPT-1B,...
2025.12
11.72
MIDUS-HML
Training Stage=SFT-1B,...
2025.12
11.75
OpT-DeUS
Training Stage=SFT-1B,...
2025.12
11.82
Avg-DeUS
Training Stage=CPT-1B,...
2025.12
12.04
LESA
Training Stage=CPT-1B,...
2025.12
12.06
Avg-DeUS
Training Stage=SFT-1B,...
2025.12
12.18
Llama Pro
Training Stage=CPT-1B,...
2025.12
12.33
LESA
Training Stage=SFT-1B,...
2025.12
12.37
Llama Pro
Training Stage=SFT-1B,...
2025.12
12.55
Base
Training Stage=SFT-1B,...
2025.12
13.09
Base
Training Stage=CPT-1B,...
2025.12
13.22
SOLAR
Training Stage=SFT-1B,...
2025.12
13.23
SOLAR
Training Stage=CPT-1B,...
2025.12
13.41
Feedback
Search any
task
Search any
task
