List your company

General Knowledge Reasoning on MMLU-Redux (test)

0.923HypervolumeRADAR
0.7203040.7729270.825550.878173Sep 29, 2025

Evaluation results

MethodLinks
RADAREvaluation Protocol=ID (In-distribution)2025.090.923
IRT-RouterEvaluation Protocol=ID (In-distribution)2025.090.9117
RouterBenchEvaluation Protocol=ID (In-distribution)2025.090.9053
Random-PairEvaluation Protocol=ID (In-distribution)2025.090.7281