Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Puzzle Solving on Sokoban
Loading...
43.6
Success Rate
Uniform
16.768
23.734
30.7
37.666
May 14, 2026
Success Rate
Updated 2mo ago
Evaluation Results
Method
Method
Links
Success Rate
Uniform
Algorithm=PPO, Model S...
2026.05
43.6
ActFocus
Algorithm=PPO, Model S...
2026.05
40.4
ActFocus
Algorithm=PPO, Model S...
2026.05
37.5
Uniform
Algorithm=PPO, Model S...
2026.05
28.7
ActFocus
Algorithm=GRPO, Model...
2026.05
23.4
ActFocus
Algorithm=GRPO, Model...
2026.05
23.2
Uniform
Algorithm=GRPO, Model...
2026.05
20.7
Uniform
Algorithm=GRPO, Model...
2026.05
17.8
Feedback
Search any
task
Search any
task
