List your company

Model Ranking Prediction on Helpsteer 30B+ Models Holdout (test)

76.5Pairwise Accuracy (RM1)BENCHALIGN
54.24460.02265.871.578Feb 2, 2026

Evaluation results

MethodLinks
BENCHALIGNHoldout=30B+ Models2026.0276.563.70.710.387
TINYBENCHMARKSHoldout=30B+ Models2026.0260.849.10.324-0.013
METABENCHHoldout=30B+ Models2026.0260.649.70.3280.007
RANDOMHoldout=30B+ Models2026.0255.145.20.108-0.148