0

Reward Bench

Evaluates pair-wise answers from RewardBench datasets

Type
RL Env
Capabilities
Safety
Runtime
single-turn
License
unknown
Size
v0.1.1
Published
Oct 2025
Updated
Oct 2025

Cite

Notes

Only stored in your browser.

Lift evidence

2
EvalTools known to liftSource paper
RewardBenchReward Bench-
RewardBench 2Reward Bench-