Complex Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
90.5Accuracy
99
Jun 5, 2026
65.1Accuracy
81
Jun 4, 2026
2.188Perplexity
44
Mar 4, 2026
65.81Accuracy
42
Feb 26, 2026
46.5Accuracy
39
Jun 23, 2026
76.17Macro-F1
37
Jun 2, 2026
78.4Macro-F1
37
Jun 2, 2026
85.18Macro F1
37
Jun 2, 2026
89.14Accuracy (%)
28
Jun 5, 2026
86.02Rel
27
Jun 30, 2026
53.1Accuracy
24
Jun 23, 2026
76.4Accuracy
19
Apr 21, 2026
83.03Acc
16
Apr 7, 2026
74.3Score
14
Jun 4, 2026
90.6Accuracy
13
Apr 21, 2026
18.4Pass@1 Score
13
Mar 4, 2026
1.64Decoding Speedup
12
Jun 30, 2026
52.9Overall Accuracy
12
May 26, 2026
28.35Accuracy
12
May 12, 2026
75.03Accuracy (ACC)
9
Jun 30, 2026
38.1Accuracy
9
Mar 19, 2026
87.3Accuracy
8
Jul 3, 2026
28.6Accuracy
8
Jul 3, 2026
91.89Recall
8
Jun 17, 2026
90.42Recall
8
Jun 2, 2026