Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Manipulation on Guava Simulation (OOD Prompt)
Loading...
100
Stack Cube (Reverse Order) Success Rate
GPT-5.4
16.8
38.4
60
81.6
Jun 16, 2026
Stack Cube (Reverse Order) Success Rate
Success Rate
Arrange by Size (Reverse) Success Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Stack Cube (Reverse Order) Success Rate
Success Rate
Arrange by Size (Reverse) Success Rate
GPT-5.4
Episodes=15
2026.06
100
-
26.7
Guava-Agent-4B
Episodes=15
2026.06
100
-
66.7
CaP-Agent0
Episodes=15
2026.06
60
-
46.7
Qwen3.5-4B
Episodes=15
2026.06
20
-
13.3
Feedback
Search any
task
Search any
task