Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Code Generation on MBPP (Acc., TPS)

56.4Accuracy

DSB Cache (greedy)

23.95232.37640.849.224Feb 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
56.461.85
2026.02
5655.67
2026.02
55.860.54
2026.02
55.462.03
2026.02
55108.5
2026.02
54.8115.3
2026.02
54.213.54
2026.02
5480.3
2026.02
53.880.76
2026.02
53.876.9
2026.02
53.857.94
2026.02
5329.01
2026.02
53101.1
2026.02
5259.48
2026.02
39.25.32
2026.02
39.231.08
2026.02
3747.61
2026.02
36.630.58
2026.02
34.829.62
2026.02
34.250.61
2026.02
3448.04
2026.02
3153.56
2026.02
30.853.35
2026.02
29.652.72
2026.02
29.414.33
2026.02
27.882.02
2026.02
27.481.88
2026.02
25.276.61