Code Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
100Easy Accuracy
19
Mar 10, 2026
123.1Retention (%)
19
Mar 10, 2026
100Retention (%)
19
Mar 10, 2026
64.1Speedup Ratio
19
Jun 1, 2026
83.6MBPP
19
Feb 26, 2026
79.5Average Pass@1
19
Feb 26, 2026
67.2Matplotlib
19
Feb 26, 2026
65.61Accuracy
18
Jun 30, 2026
81.71Accuracy
18
Jun 30, 2026
71.2Accuracy
18
Jun 23, 2026
95.1Accuracy
18
Jun 23, 2026
84.5Score
18
Jun 23, 2026
94.5Pass@1
18
Jun 17, 2026
87.4Pass@1
18
Jun 16, 2026
46.8Pass@1
18
Jun 16, 2026
65.5Code Avg
18
Jun 5, 2026
79.2Average Accuracy
18
Jun 2, 2026
64.6Average Accuracy
18
Jun 2, 2026
66.7Average Accuracy
18
Jun 2, 2026
48.2Pass@1 (HumanEval)
18
Jul 7, 2026
14.8ΔPPL (%)
18
May 29, 2026
69.6TS
18
May 26, 2026
6.17Speed-up
18
May 15, 2026
98.6Easy Success Rate
18
May 13, 2026
57.2Score (%)
18
May 13, 2026