Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Sudoku Solving on 9x9 Sudoku (test)
Loading...
52
Cell Accuracy
Fine-tuned (solver order)
25.168
32.134
39.1
46.066
Dec 3, 2025
Cell Accuracy
FSR
GPA
Updated 1mo ago
Evaluation Results
Method
Method
Links
Cell Accuracy
FSR
GPA
Fine-tuned (solver order)
Training Protocol=Supe...
2025.12
52
-
-
GRPO with Bootstrapped Mixed Rewards
Cell : Order Weight=0....
2025.12
49.6
-
-
GRPO with Bootstrapped Mixed Rewards
Cell : Order Weight=0....
2025.12
46.5
-
-
GRPO with Bootstrapped Mixed Rewards
Cell : Order Weight=1...
2025.12
43.8
-
-
GRPO with Bootstrapped Mixed Rewards
Cell : Order Weight=0....
2025.12
37.8
-
-
Fine-tuned (random order)
Training Protocol=Supe...
2025.12
28.2
-
-
GRPO with Bootstrapped Mixed Rewards
Cell : Order Weight=0...
2025.12
26.2
-
-
HRM
Training Data=9x9 Sudo...
2026.03
-
63.53
86.11
TRM
Training Data=9x9 Sudo...
2026.03
-
71.94
89.8
SE-RRM
Training Data=9x9 Sudo...
2026.03
-
93.73
97.58
GPT-OSS-20B
Evaluation protocol=Wi...
2026.03
-
21.67
-
Feedback
Search any
task
Search any
task