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

Code Generation on MBPP (Acc. %, Steps, Speed)

58Accuracy

COVER

35.32841.21447.152.986Feb 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
5842.354.36
2026.02
57.635.8411.64
2026.02
57.25121
2026.02
57207.171.31
2026.02
5757.084.23
2026.02
56.82561
2026.02
56.262.117.82
2026.02
55.4461.571.22
2026.02
42.475.372.42
2026.02
41.882.081.55
2026.02
41.2107.112.34
2026.02
4194.033.82
2026.02
40.82561
2026.02
40.4125.351.92
2026.02
40.2167.993.05
2026.02
39.85121
2026.02
39.4123.482.83
2026.02
3969.752.35
2026.02
38.8115.243.08
2026.02
38.62561
2026.02
38.4131.291.94
2026.02
38.4185.52.7
2026.02
38.25121
2026.02
38139.511.84
2026.02
37.889.991.65
2026.02
37.870.522.45
2026.02
37.6114.862.25
2026.02
37.65121
2026.02
37.6124.422.38
2026.02
37.4168.442.96
2026.02
37.22561
2026.02
36.2871.64