Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Performance Prediction on MMLU 14k (test)
Loading...
1.07
MAE
DISCO
0.9748
1.6174
2.26
2.9026
Oct 9, 2025
MAE
Rank Correlation
Updated 1mo ago
Evaluation Results
Method
Method
Links
MAE
Rank Correlation
DISCO
Selection=High PDS, Pr...
2025.10
1.07
0.987
DISCO
Selection=High JSD, Pr...
2025.10
1.14
0.975
DISCO
Selection=High JSD, Pr...
2025.10
1.3
0.987
DISCO
Selection=High PDS, Pr...
2025.10
1.31
0.972
Model signature
Selection=Random, Pred...
2025.10
1.81
0.933
Model signature
Selection=Random, Pred...
2025.10
1.82
0.912
tinyBenchmarks
Selection=Anchor-corr,...
2025.10
2.08
0.927
Metabench
Selection=Best for val...
2025.10
2.08
0.904
tinyBenchmarks
Selection=Random, Pred...
2025.10
2.79
0.922
tinyBenchmarks
Selection=Anchor-IRT,...
2025.10
3.25
0.922
Baseline
Selection=Random, Pred...
2025.10
3.45
0.916
Feedback
Search any
task
Search any
task