MBPP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MBPP+
74.5Average Success Rate (@32)
12
MBPP (test)
51.58Accuracy
12
MBPP sanitized
64.59Accuracy
12
MBPP
82.29Avg@8
12
MBPP
0.68MA
12
MBPP-Hard
62.56Pass Rate
12
MBPP New
70.13Pass Rate
12
MBPP Original
70.86Pass Rate
12
MBPP
4.78Speedup
12
MBPP
96.5Pass@1 Accuracy (Python)
12
MBPP
72.47Accuracy
11
MBPP
86.7Accuracy
11
MBPP Python
69.8Pass@1
11
MBPP
66.7MBPP Accuracy (%)
10
MBPP++
65.87Accuracy
10
MBPP+
82.8Accuracy
10
MBPP
44.73Pass@1
10
MBPP
64.59Accuracy (MBPP)
10
MBPP EvalPlus
84.9Pass@1
10
MBPP
38.5Pass@1
10
MBPP-Instruct 0-shot (test)
60.6TPS
10
MBPP 1,000-example (test)
9.021Perplexity
10
MBPP Sanitized (test)
84.3Pass@1
10
MBPP
8.58NFE Reduction
9
MBPP
93.77Score
9