Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Logic Puzzle on Sudoku (Success rate)
Loading...
99.6
Success Rate
ActFocus
-3.984
22.908
49.8
76.692
May 14, 2026
Success Rate
Updated 19d ago
Evaluation Results
Method
Method
Links
Success Rate
ActFocus
Algorithm=PPO, Model S...
2026.05
99.6
Uniform
Algorithm=PPO, Model S...
2026.05
84.2
ActFocus
Algorithm=PPO, Model S...
2026.05
82
ActFocus
Algorithm=GRPO, Model...
2026.05
80.9
Uniform
Algorithm=GRPO, Model...
2026.05
70.9
Uniform
Algorithm=PPO, Model S...
2026.05
16.2
Uniform
Algorithm=GRPO, Model...
2026.05
0
ActFocus
Algorithm=GRPO, Model...
2026.05
0
Feedback
Search any
task
Search any
task
