MBPP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MBPP Synthetic
7.9Personalization Score
6
MBPP
0.063Inference Cost (USD)
6
MBPP
56.8Pass@1 Accuracy
6
MBPP+ instruct latest (test)
75.4Accuracy
6
MBPP 3-shot pass@1
25.6Accuracy
6
MBPP
1TPF
6
MBPP Hard-Guided
72.44Pass Rate
6
MBPP Guided
77.19Pass Rate
6
MBPP
41Pass@1
6
MBPP (test)
0.459Accuracy
6
MBPP+
60.58Accuracy
6
MBPP
89Score
6
MBPP
0.37Cost
6
MBPP
62.65Accuracy
5
MBPP
45.38Accuracy
5
MBPP+
64.5Score (%)
5
MBPP → MBPP+ (mild)
4.5ECE
5
MBPP LCB (moderate)
7.5ECE
5
MBPP → CC (severe)
6.3ECE
5
MBPP → BCB (severe)
10.9ECE
5
MBPP
57.4Accuracy
5
MBPP+
64Pass Rate
5
MBPP-S
100TPR
5
MBPP Instruct
5.86TPF
5
MBPP
1Average Code Length
5