Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Pour Water on Real-robot tabletop tasks 15 rollouts
Loading...
46.7
Success Rate
SkiP
5.1
15.9
26.7
37.5
May 15, 2026
Success Rate
Steps
Time (s)
Updated 16d ago
Evaluation Results
Method
Method
Links
Success Rate
Steps
Time (s)
SkiP
Backbone=π0.5, Fine-tu...
2026.05
46.7
265.4
180
Base
Backbone=π0.5, Fine-tu...
2026.05
40
290.4
180
CoA
Backbone=π0.5, Fine-tu...
2026.05
33.3
281.9
180
KF-only
Backbone=π0.5, Fine-tu...
2026.05
6.7
309.6
240
Feedback
Search any
task
Search any
task