| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multimodal Reasoning | GeoBench-VLM | Aerial Score35.4 | 11 | |
| 2D-edits | GeoBench 1.0 (test) | FID25.07 | 9 | |
| Visual Question Answering | GEOBench VLM | Object Localization & Counting Score39.5 | 8 | |
| 3D-edits | GeoBench 1.0 (test) | FID64.3 | 4 |