List your company

Model Ranking Prediction on Helpsteer 70B+ Models Holdout (test)

77.8Pairwise Acc (RM1)BENCHALIGN
52.8459.3265.872.28Feb 2, 2026

Evaluation results

MethodLinks
BENCHALIGNHoldout=70B+ Models2026.0277.8620.7070.333
METABENCHHoldout=70B+ Models2026.0256.448.40.2-0.028
TINYBENCHMARKSHoldout=70B+ Models2026.0254.749.40.16-0.004
RANDOMHoldout=70B+ Models2026.0253.846.80.087-0.106