Code Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
94.3AUCOAA
11
Feb 26, 2026
41.47ATS
11
Feb 26, 2026
86.7Accuracy
11
Feb 26, 2026
90.5Pass@1
11
Feb 26, 2026
69.8Pass@1
11
Feb 26, 2026
70.4Accuracy
10
Jul 1, 2026
41.4Accuracy
10
Jul 1, 2026
66.7MBPP Accuracy (%)
10
Jun 19, 2026
0.543CodeBLEU
10
Jun 16, 2026
58.8Accuracy
10
Jun 16, 2026
58.54Accuracy (%)
10
Jun 8, 2026
2.18Speedup
10
Jun 4, 2026
79.2Accuracy
10
Jun 2, 2026
88.9MBPP Score
10
May 29, 2026
47.14Avg@4
10
May 27, 2026
11Pass@1
10
May 19, 2026
19.6pass@1
10
May 19, 2026
80.1Expected Utility
10
May 14, 2026
61.5Pass Rate
10
May 13, 2026
96.51S_ad
10
May 12, 2026
31.3Accuracy
10
May 12, 2026
265.52Decoding Speed (tokens/s)
10
May 4, 2026
82.8Accuracy
10
May 11, 2026
81.5Accuracy
10
Apr 24, 2026
11Pass@1
10
Apr 16, 2026