MBPP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MBPP+
0.738Accuracy
9
MBPP (test)
78.42AUROC
9
MBPP
0.429nR2
9
MBPP+
54.8Pass@1
9
MBPP (test)
74.3Pass@1
9
MBPP 256
57.6Accuracy
9
MBPP OOD Generalization (trained on CodeContests)
85.45Pass@128
9
MBPP
81.2Accuracy
9
MBPP
59.03Accuracy
9
MBPP Pro
85.56Pass@1
9
MBPP
284TPF
9
MBPP
77.49MBPP avg@16 Accuracy
9
MBPP
0.406Accuracy
9
MBPP
5.26Time per Token (ms/tok)
8
MBPP
5.64Latency (ms/tok)
8
MBPP+ (F1)
99.32Release Rate
8
MBPP+ (F0)
0False Release Rate
8
MBPP
44.11Pass@1
8
MBPP-G
100Message Accuracy
8
MBPP Sanitized
85.97pass@1
8
MBPP
0.37Top-1 ECE
8
MBPP
53.7pass@16 (MBPP)
8
MBPP
55.6Accuracy
8
MBPP
29.9Answer Correctness Rate
8
MBPP 1-shot
48.46Pass@16
8