Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Continuous Control (MuJoCo) on Hopper v4
Loading...
3,240
Mean Episodic Return
FP16-PPO
2,740.8
2,870.4
3,000
3,129.6
Apr 27, 2026
Mean Episodic Return
FP16 Usage Percentage
Updated 1mo ago
Evaluation Results
Method
Method
Links
Mean Episodic Return
FP16 Usage Percentage
FP16-PPO
Precision=FP16
2026.04
3,240
-
INT8-PPO
Precision=INT8
2026.04
3,180
-
INT4-PPO
Precision=INT4
2026.04
3,050
-
BitRL-1bit
Precision=1-bit
2026.04
2,890
89.2
MLP-PPO
Architecture=MLP
2026.04
2,760
-
Feedback
Search any
task
Search any
task