Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
52.2Accuracy (Test-All)
7
Feb 26, 2026
55.8Test-All Score
7
Feb 26, 2026
52.7Accuracy (All Test Cases)
7
Feb 26, 2026
78.6Pass@1
6
Jun 30, 2026
36.9Pass@1
6
Jun 25, 2026
34.92Accuracy (LogiQA CN)
6
Jun 17, 2026
100Accuracy
6
Jun 16, 2026
96.3Accuracy (Avg@8)
6
Jun 5, 2026
17Accuracy
6
May 29, 2026
73.53Static Accuracy
6
May 26, 2026
79.3Accuracy
6
May 21, 2026
92.2Acc
6
May 13, 2026
100Accuracy
6
May 7, 2026
52.5Exact Match Accuracy (BBH Logical Reasoning)
6
May 12, 2026
4.5Romi Score
6
Mar 26, 2026
98.5Accuracy
6
Apr 2, 2026
43.57Selection Accuracy
6
Feb 26, 2026
42.24Accuracy
6
Feb 26, 2026
58.4Pass@1
5
Jun 17, 2026
34.6Pass@1
5
Jun 17, 2026
46.5Accuracy
5
Jun 8, 2026
99.9Accuracy
5
Jun 4, 2026
70Accuracy
5
May 29, 2026
55.3Accuracy
5
May 29, 2026
61Accuracy
5
May 29, 2026