Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
78.9LogiQA Accuracy
251
May 19, 2026
100Accuracy
249
May 27, 2026
81.4Accuracy
163
Jul 7, 2026
86Accuracy
153
Jun 15, 2026
94.3Accuracy
152
Jun 19, 2026
87.8Accuracy
136
May 27, 2026
89.2Accuracy
126
May 11, 2026
83.8Accuracy
116
Apr 30, 2026
80.4Accuracy
106
Jun 17, 2026
50.23Accuracy
98
Feb 26, 2026
97.7Cell Accuracy
90
May 29, 2026
92.2Grid Accuracy
90
May 29, 2026
80.6Accuracy
87
Feb 26, 2026
98.8Accuracy
86
Jun 30, 2026
83.7Accuracy
80
Feb 26, 2026
0.723Accuracy
62
Apr 23, 2026
91.7Accuracy
60
Apr 20, 2026
95.6Accuracy
60
Jun 15, 2026
89Accuracy
58
Feb 26, 2026
99.72Accuracy
57
May 26, 2026
92.32Accuracy
57
May 26, 2026
88.1Accuracy
56
Mar 25, 2026
93.8Accuracy
56
Mar 25, 2026
89.5Accuracy
56
Mar 25, 2026
88.9Accuracy
55
Feb 26, 2026