Code Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
44.73Pass@1
10
Apr 16, 2026
45.12Pass@1
10
Apr 16, 2026
1.37Total Tokens
10
Apr 2, 2026
83.3Last SR
10
Apr 1, 2026
60.6Pass@1
10
Mar 16, 2026
1.41tau
10
Mar 11, 2026
4.18tau
10
Mar 11, 2026
5.65Tau
10
Mar 11, 2026
64.59Accuracy (MBPP)
10
Mar 9, 2026
59.76Accuracy
10
Mar 9, 2026
84.9Pass@1
10
Mar 4, 2026
62.5Correction Uplift (%)
10
Mar 4, 2026
38.5Pass@1
10
Mar 4, 2026
55.4Pass@1
10
Mar 4, 2026
60.6TPS
10
Feb 26, 2026
36.6Code Score
10
Feb 26, 2026
99.9Avg Executable Rate @8 (E.R.)
10
Feb 26, 2026
87.8Accuracy
10
Feb 26, 2026
21.69Accuracy
10
Feb 26, 2026
61.13Accuracy
10
Feb 26, 2026
54.27Accuracy
10
Feb 26, 2026
84.3Pass@1
10
Feb 26, 2026
44.9Java
10
Feb 26, 2026
31.98CodeBLEU
10
Feb 26, 2026
93.3Valid Rate
10
Feb 26, 2026