Code Generation On Taco Code
評価指標
easy pass@1
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | easy pass@1 | Paper Title | Repository |
|---|---|---|---|
| CodeLlama-7B-Python | 9.32% | TACO: Topics in Algorithmic COde generation dataset | - |
| Starcoder-15.5B | 11.6% | TACO: Topics in Algorithmic COde generation dataset | - |
| GPT-4 | 31.50% | TACO: Topics in Algorithmic COde generation dataset | - |
0 of 3 row(s) selected.