Code Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
87.4Accuracy
102
Jun 16, 2026
95.73HumanEval Score
70
Jun 18, 2026
83.5Accuracy
61
Jun 1, 2026
98.8Input-CoT Accuracy
56
Mar 18, 2026
84.7MBPP Execution Accuracy
39
Jun 18, 2026
76.87Accuracy
36
May 27, 2026
84.9Accuracy
35
Feb 26, 2026
53.6Accuracy
26
May 26, 2026
62.6Pass@1
26
Jun 18, 2026
94.55Accuracy
26
May 20, 2026
62.46pass@1
26
May 22, 2026
87.37Accuracy
26
May 12, 2026
74.5Pass@4
25
May 20, 2026
47.4Accuracy
24
Apr 21, 2026
49.91Accuracy
24
Apr 21, 2026
10.34Accuracy
24
Apr 21, 2026
51.1Pass@1 Accuracy
20
Apr 28, 2026
46.34Pass@1
18
Jul 7, 2026
72.29HumanEval
18
Jun 8, 2026
79.88HEval+ (Pass@1)
18
Apr 21, 2026
47.2Accuracy
18
Feb 26, 2026
1.2VT Score
16
Jun 17, 2026
33.72Accuracy
16
May 26, 2026
97Pass@16
15
Jun 17, 2026
43.2Avg@16
15
Jun 17, 2026