Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language-Conditioned Tasks on SmartHome Light
Loading...
0.85
Mean Episodic Return
FP16-PPO
0.7044
0.7422
0.78
0.8178
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
0.85
-
INT8-PPO
Precision=INT8
2026.04
0.84
-
INT4-PPO
Precision=INT4
2026.04
0.82
-
BitRL-1bit
Precision=1-bit
2026.04
0.82
96.5
MLP-PPO
Architecture=MLP
2026.04
0.71
-
Feedback
Search any
task
Search any
task