Logical reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
15.5LRL Score
54
May 12, 2026
95.9Accuracy
54
Jun 24, 2026
58.37Accuracy
50
Feb 26, 2026
64.9Avg Pass@8
48
Apr 15, 2026
0.786Accuracy
46
Feb 26, 2026
98.4Accuracy
44
Mar 24, 2026
75Accuracy
43
May 26, 2026
47.3Accuracy
40
Feb 26, 2026
85.43Accuracy
39
Mar 19, 2026
80.1Accuracy
39
Jun 15, 2026
92.1Accuracy
39
May 14, 2026
87.2Accuracy
37
Feb 26, 2026
60.22Accuracy
34
May 14, 2026
94.1Accuracy (Avg@8)
32
Jun 30, 2026
2.13Perplexity (3ppl)
32
May 26, 2026
78Accuracy
32
Mar 4, 2026
94.33Accuracy
31
Apr 21, 2026
63.2Accuracy
31
Jun 19, 2026
100Accuracy
29
May 15, 2026
88.29Top@1 Accuracy
29
Apr 23, 2026
81.8Pass@1 Rate (4-8 Roles, ID)
28
Jun 11, 2026
47.5Test Accuracy
28
Jun 2, 2026
67.7Accuracy
28
Apr 23, 2026
80.4Accuracy
28
Apr 20, 2026
86.4Accuracy
28
Feb 26, 2026