Property Damage Extent nowcasting on Harris County (test)
71.37Overall AccuracyFloodDamageCast*
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| FloodDamageCast*2026.02 | 71.37 | 68.49 | 85.9 | 84.36 | 67.21 | 3 | 28.6 | |
| gpt-4.1Temp.=0.52026.02 | 66.75 | 63.51 | 81.62 | 82.51 | 43.92 | 13.3 | 6.2 | |
| gpt-4o-miniTemp.=0.52026.02 | 66.02 | 61.01 | 77.75 | 81.92 | 34.57 | 1 | 81.9 | |
| gpt-4oTemp.=0.52026.02 | 65.97 | 61.74 | 80.42 | 82.04 | 38.28 | 16.7 | 4.9 | |
| gpt-5-miniTemp.=1.02026.02 | 64.72 | 62.84 | 86.75 | 80.85 | 59.1 | 2.3 | 35.9 | |
| deepseek-r1Temp.=0.52026.02 | 64.42 | 60.14 | 75.7 | 80.05 | 45.83 | 5 | 16 | |
| qwen3 (30B-A3B-instruct-2507)Temp.=0.52026.02 | 64.19 | 62.38 | 81.14 | 80.72 | 57.64 | 13.2 | 6.1 | |
| llama-3.1 (70B-instruct)Temp.=0.52026.02 | 61.28 | 58.51 | 89.56 | 78.73 | 52.94 | 19.9 | 4 |