Logic reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
71.33Accuracy
59
May 20, 2026
96Accuracy
49
Jun 9, 2026
96.1Score
42
Feb 26, 2026
36Accuracy
30
Feb 26, 2026
99Score Threshold 2 (OOD)
25
Mar 10, 2026
98Accuracy (Level 3)
25
Mar 10, 2026
61.97LogicVista Accuracy
16
Jun 1, 2026
0.439Score
16
Feb 26, 2026
40.3Score
16
Feb 26, 2026
0.817Avg Accuracy @1
11
Feb 26, 2026
75Tasks Solved
10
Jun 2, 2026
26Tasks Solved
9
Jun 2, 2026
71.7Accuracy
9
Apr 14, 2026
67.9MMStar Score
8
May 18, 2026
98.7Accuracy
7
May 7, 2026
69F1 Score
7
Mar 6, 2026
41F1 Score
7
Mar 6, 2026
19F1 Score
7
Mar 6, 2026
49.2Reasoning Score
6
Jun 11, 2026
54Accuracy on LogicVista
6
May 6, 2026
77Pass@8
6
Apr 10, 2026
28Visu Logic Score
4
May 6, 2026
38.31ARC-C Accuracy
4
Feb 26, 2026
95.6Accuracy
3
Feb 26, 2026
69.8Pass@1
3
Feb 26, 2026