Python
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Python (test)
94.4Safe Code Rate
66
Python 13
83.4mAP
56
Python
7.69MAT
30
Python (test)
49.4Recall@1
25
Python
0.497Pearson Correlation (rp)
15
Python (Py)
99.4Decoupling Probability
9
Python deduplicated retrieval codebase (test)
1,183EM
9
Python (test)
21.01BLEU
8
Python 10 sequential 20-step campaigns (end-of-chain)
25.9End pass@1
5
Python
85.63MRR
5
Python (held-out eval)
37.7Pass@1
4
Python Full (val)
92.4Precision
4
Python 2D
4.4Wall-Clock Time (ms)
2