Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Reward Modeling on RewardBench 2

93.4L-Acc

GPT-5-chat

82.27285.16188.0590.939Feb 6, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
93.410.383.7
2026.02
93.15.188.3
2026.02
92.63.489.4
2026.02
921.790.4
2026.02
9214.678.5
2026.02
91.918.974.5
2026.02
91.69.483
2026.02
90.616.275.9
2026.02
90.213.977.7
2026.02
90.14.885.7
2026.02
89.821.770.3
2026.02
89.225.966.1
2026.02
88.536.756
2026.02
88.12962.5
2026.02
87.920.170.2
2026.02
87.819.370.9
2026.02
87.118.371.2
2026.02
85.327.461.9
2026.02
84.22563.2
2026.02
82.724.662.4