codesearchnet---java is a state-of-the-art machine learning benchmark indexed on Codesota. This page tracks published model results, top scores per metric, and the SOTA timeline for codesearchnet---java.
Smoothed Bleu 4 is the reported evaluation metric for codesearchnet---java. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
| Rank | Model | Trust | Score | Year | Links | Fix |
|---|---|---|---|---|---|---|
| 01 | StarCoder-LoRA | unverified | 22.61 | 2024 | Source ↗ | Looks wrong? |
| 02 | CodeTrans-MT-Large | verified | 21.87 | 2021 | Paper ↗Code ↗ | Looks wrong? |
| 03 | DistillCodeT5 | unverified | 20.51 | 2024 | Source ↗ | Looks wrong? |
| 04 | PolyglotCodeBERT | verified | 20.11 | 2021 | Source ↗ | Looks wrong? |
| 05 | ProphetNet-X | verified | 19.39 | 2021 | Source ↗ | Looks wrong? |
| 06 | CoTexT | verified | 19.06 | 2021 | Source ↗ | Looks wrong? |
| 07 | PLBART | verified | 18.45 | 2021 | Source ↗ | Looks wrong? |
| 08 | CodeBERT (MLM+RTD) | verified | 14.56 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 09 | CodeBERT (MLM) | verified | 13.59 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 10 | RoBERTa | verified | 13.2 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 11 | pre-train w/ code only | verified | 13.07 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 12 | CodeBERT (RTD) | verified | 12.72 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 13 | Transformer | verified | 12.57 | 2020 | Paper ↗Code ↗ | Looks wrong? |
| 14 | seq2seq | verified | 11.42 | 2020 | Paper ↗Code ↗ | Looks wrong? |