Code Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
58.65Pass@1
28
Feb 26, 2026
71.98MBPP Performance Score
28
Mar 17, 2026
83Accuracy
28
Feb 26, 2026
26.9Accuracy
27
Jun 11, 2026
35.88Variation Rate @ k=10
27
Jun 5, 2026
84.07Average Score (AVG)
27
Jun 11, 2026
18.37Avg@5 Accuracy
27
Apr 29, 2026
81.1Accuracy
27
Mar 17, 2026
98.78Accuracy
27
Apr 28, 2026
34.2HumanEval
26
Jun 11, 2026
47.6MBPP Accuracy
26
Jun 11, 2026
4.15Inference Speed (× Baseline)
26
May 26, 2026
34.1Sig. Score
26
May 25, 2026
15.3Sig. Score
26
May 25, 2026
56.5Sig. Score
26
May 25, 2026
8.53MAT Score
26
May 20, 2026
90.8Accuracy
26
Jun 30, 2026
3.934Speedup Ratio
26
Apr 8, 2026
39.33Avg@8
26
Feb 26, 2026
76.9Pass@1
26
Jul 7, 2026
63.9Pass@1
26
May 7, 2026
90.2Score
26
Feb 26, 2026
0.879BLEU
26
Feb 26, 2026
351.07TPS (Tokens/s)
25
May 4, 2026
92.1pass@1
25
May 28, 2026