Polygon Area Estimation on SpatialEval Los Angeles 1.0 (test)
884.4MAESRL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SRL2026.06 | 884.4 | — | |
| ChatGPT 5.1mode=In-Context2026.06 | 1,137.8 | 100 | |
| SLMbase_model=Qwen3-8B, fine-tuning=LoRA2026.06 | 1,175.3 | 100 | |
| ChatGPT 5.12026.06 | 1,291.6 | 100 | |
| gemini-2.5-flash2026.06 | 1,562.4 | 100 | |
| Qwen3-8Bmode=tuned2026.06 | 1,687.1 | 100 | |
| SpatialRGPT2026.06 | 1,698.8 | 100 | |
| Qwen3-8B2026.06 | 5,053.4 | 97.7 | |
| DeepSeek-R1-32B2026.06 | 8,979.6 | 37.5 | |
| Llama-3.3-70B2026.06 | 815,900.9 | 82.8 |