Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Reward Modeling on RewardBench latest (full)

93.6Average Score

BNBT-Reward-Llama-3.1-8B

86.42488.28790.1592.013Feb 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
93.695.389.792.696.9
2026.02
92.796.984.891.697.6
2026.02
92.595.887.390.696.5
2026.02
92.295.887.192.293.6
2026.02
91.894.684.591.596.5
2026.02
90.896.976.892.297.3
2026.02
90.397.282.286.595.1
2026.02
90.298.976.589.995.8
2026.02
89.497.281.886.891.9
2026.02
86.894.17785.890.2
2026.02
86.796.176.188.186.6