Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reward Prediction on OSWorld
Loading...
92
Reward Accuracy
PRORE
80.56
83.53
86.5
89.47
Sep 26, 2025
Reward Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
Reward Accuracy
PRORE
Evaluator Agent=Claude...
2025.09
92
DigiRL
Evaluator Agent=Claude...
2025.09
88
DistRL
Evaluator Agent=Claude...
2025.09
88
WebRL
Evaluator Agent=Claude...
2025.09
86
StepCritic
Evaluator Agent=Claude...
2025.09
81
Feedback
Search any
task
Search any
task