Puzzle Solving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
27.2Test Accuracy
42
Jun 9, 2026
95.8Accuracy
22
Feb 27, 2026
0.847Accuracy
20
Feb 27, 2026
29Score
12
Apr 21, 2026
43Score
12
Apr 21, 2026
42.6Avg@128
12
Mar 23, 2026
66.9Avg@128
12
Mar 23, 2026
77.8Average @128
12
Mar 23, 2026
79.8Avg@128
12
Mar 23, 2026
89.7Average Score @128
12
Mar 23, 2026
97.1Average Score @128
12
Mar 23, 2026
23Success Rate (Offline)
10
Apr 21, 2026
94.2Pass@1 Success Rate
9
Jun 17, 2026
43.6Success Rate
8
May 15, 2026
30PA
8
May 13, 2026
28.5PA
8
May 13, 2026
97.2Puzzle Success Rate
7
May 20, 2026
93.1Pass@1
6
Jun 17, 2026
49Solve Rate
5
Feb 26, 2026
100Admissibility
4
Jun 4, 2026
100Admissibility
4
Jun 4, 2026
100Admissibility Rate
4
Jun 4, 2026
54.3Accuracy
4
Jun 4, 2026
5.7Accuracy
4
Jun 4, 2026
58.1Accuracy
4
Jun 4, 2026