Visual Grounding on VRSBench
64.57mIoUGeoSolver
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| GeoSolver2026.03 | 64.57 | 64.57 | — | |
| DiffuSAMTask-specific training=None2026.04 | 64.14 | 79.13 | 40.51 | |
| RemoteSAMTask-specific training=None2026.04 | 62.83 | 78.12 | 39.31 | |
| EarthMindTask-specific training=None2026.04 | 59.61 | 74.14 | 37.32 | |
| FalconTask-specific training=None2026.04 | 58.02 | 72.21 | 47.09 | |
| GLM-4.1V-ThinkingModel Category=Open-source Reasoning Vision-Language Models2026.03 | 43.2 | 43.2 | — | |
| VHMModel Category=Open-source Remote Sensing Vision-Language Models2026.03 | 34.91 | 34.91 | — | |
| GeoChatModel Category=Open-source Remote Sensing Vision-Language Models2026.03 | 23.66 | 23.66 | — | |
| SkySenseGPTModel Category=Open-source Remote Sensing Vision-Language Models2026.03 | 21.25 | 21.25 | — | |
| SAM3Task-specific training=None2026.04 | 16.35 | 20.51 | 8.87 | |
| Gemini-2.0-flashModel Category=Closed-source Commercial Vision-Language Models2026.03 | 13.52 | 13.52 | — | |
| EarthDialModel Category=Open-source Remote Sensing Vision-Language Models2026.03 | 13.04 | 13.04 | — | |
| Claude-sonnet-4Model Category=Closed-source Commercial Vision-Language Models2026.03 | 6.65 | 6.65 | — | |
| ChatGPT-5Model Category=Closed-source Commercial Vision-Language Models2026.03 | 5.65 | 5.65 | — | |
| RS-EoTModel Category=Open-source Reasoning Vision-Language Models2026.03 | 1.28 | 1.28 | — |