Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Continuous Control (MuJoCo) on Walker2d v4
Loading...
4,120
Mean Episodic Return
FP16-PPO
3,423.2
3,604.1
3,785
3,965.9
Apr 27, 2026
Mean Episodic Return
FP16 Usage (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Mean Episodic Return
FP16 Usage (%)
FP16-PPO
Precision=FP16
2026.04
4,120
-
INT8-PPO
Precision=INT8
2026.04
4,050
-
INT4-PPO
Precision=INT4
2026.04
3,890
-
BitRL-1bit
Precision=1-bit
2026.04
3,620
87.9
MLP-PPO
Architecture=MLP
2026.04
3,450
-
Feedback
Search any
task
Search any
task