Codebase generation on Financial Tracking App
62Feature CompletenessCode-L2MAC
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Code-L2MAC2023.10 | 62 | 0 | 307 | 12 | 90.5 | |
| GPT42023.10 | 26.2 | 0.0513 | 80.5 | 2.13 | 93.1 | |
| AutoGPT2023.10 | 25 | 0 | 0 | 0 | 0 | |
| Self-Refine2023.10 | 23.6 | 0.25 | 87.2 | 0.55 | 76.7 | |
| Reflexion2023.10 | 22.5 | 0.2 | 86.8 | 2.7 | 92.8 | |
| CodeT2023.10 | 21.4 | 0 | 65.9 | 2.25 | 97.9 |