Spatial Reasoning on CV-Bench (test)
83.62D ScoreGCA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GCAParadigm=Tool-Integrated Spatial Agents2025.11 | 83.6 | 90.8 | 86.9 | |
| Qwen3-VL-ThinkingParadigm=Baseline Foundation VLMs2025.11 | 81.9 | 92.6 | 86.8 | |
| Gemini-2.5-ProParadigm=Baseline Foundation VLMs2025.11 | 81.2 | 92.5 | 86.3 | |
| GLM-4.5VParadigm=Baseline Foundation VLMs2025.11 | 80.7 | 91.6 | 85.6 | |
| VLaserParadigm=Training-based Spatial VLMs2025.11 | 79.9 | 87.8 | 83.6 | |
| RoboBrain-2.0Paradigm=Training-based Spatial VLMs2025.11 | 77.1 | 90.7 | 83.4 | |
| VILASRParadigm=Training-based Spatial VLMs2025.11 | 75.7 | 77.7 | 76.6 | |
| TIGeRParadigm=Tool-Integrated Spatial Agents2025.11 | 75.2 | 95.7 | 84.5 | |
| SpaceRParadigm=Training-based Spatial VLMs2025.11 | 74.1 | 77.4 | 75.6 | |
| Video-R1Paradigm=Training-based Spatial VLMs2025.11 | 73.5 | 74.7 | 74 | |
| SpatialLadderParadigm=Training-based Spatial VLMs2025.11 | 72.4 | 74.9 | 73.7 | |
| GPT-4oParadigm=Baseline Foundation VLMs2025.11 | 69.4 | 84.9 | 76.5 | |
| Spatial-MLLMParadigm=Training-based Spatial VLMs2025.11 | 59.5 | 63.3 | 61.2 | |
| SpatialLLMParadigm=Training-based Spatial VLMs2025.11 | 51.3 | 78.6 | 64.5 |