Iterative model calibration on VIC hydrological model Bengbu basin 1980-1990
100Success Rate (>= 0.7)Claude Opus 4.5
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Claude Opus 4.5Model=Claude Opus 4.52026.05 | 100 | 0.706 | 0.729 | 0.008 | 34.1 | 128 | |
| Codex 5.3Model=Codex 5.32026.05 | 100 | 0.703 | 0.707 | 0.003 | 21.5 | 74 | |
| Codex 5.2Model=Codex 5.22026.05 | 70 | 0.697 | 0.719 | 0.027 | 20.8 | 70 | |
| Gemini 3 FlashModel=Gemini 3 Flash2026.05 | 50 | 0.697 | 0.716 | 0.011 | 40.8 | 135 | |
| Claude Sonnet 4.5Model=Claude Sonnet 4.52026.05 | 40 | 0.675 | 0.704 | 0.029 | 34.4 | 117 | |
| Kimi K2.5Model=Kimi K2.52026.05 | 10 | 0.602 | 0.708 | 0.056 | 3.9 | 19 |