MBPP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MBPP
38NFE (Number of Failed Executions)
2
MBPP
97.85Bit Accuracy
2
MBPP-G
71.28Pass Rate Before
2
MBPP
69.8Accuracy
2
MBPP
0.58Inference Cost (USD)
2
MBPP-plus
48.7Pass@1
2
MBPP
720Throughput (tok/s)
2
MBPP
65.76Score
2
MBPP Rust MultiPL-E translation (test)
75.4Pass@1
2
MBPP
60Winrate
1
MBPP
—Pass@1
0