Spatial Reasoning on CVBench (Count, 2DRel, 3DDep, 3DDis, Avg Metrics)
70.6Count ScoreEmbodied-R1
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Embodied-R1Model Category=Open-source models, Rank=2.12025.08 | 70.6 | 90.8 | 84.7 | 84.8 | 82.7 | |
| Embodied-R1 w/o CSModel Category=Open-source models, Rank=3.42025.08 | 70.4 | 90.2 | 84.5 | 81 | 81.5 | |
| Qwen2.5VL-3BModel Category=Open-source models, Rank=5.62025.08 | 68.4 | 72.8 | 77 | 68.2 | 71.6 | |
| Embodied-SFTModel Category=Open-source models, Rank=3.72025.08 | 66.4 | 92.3 | 85.8 | 83.8 | 82.1 | |
| GPT4oModel Category=Closed-source models2025.08 | 65.9 | 85.5 | 87.8 | 78.2 | 79.4 | |
| RoboBrain-7BModel Category=Open-source models, Rank=4.42025.08 | 64.3 | 76.6 | 84 | 72 | 74.2 | |
| GPT4VModel Category=Closed-source models2025.08 | 62.4 | 71.1 | 79.8 | 68.3 | 70.4 | |
| FSD-13BModel Category=Open-source models, Rank=4.62025.08 | 62.4 | 86.5 | 88 | 86.7 | 80.9 | |
| SAT-Dynamic-13BModel Category=Open-source models, Rank=6.62025.08 | 61.5 | 89.7 | 80.7 | 73 | 76.2 | |
| ASMv2-13BModel Category=Open-source models, Rank=7.22025.08 | 58.9 | 68.9 | 68.9 | 68.9 | 66.4 | |
| LLaVA-1.5-13BModel Category=Open-source models, Rank=9.42025.08 | 58.2 | 46.6 | 53 | 47.8 | 51.4 | |
| RoboPoint-13BModel Category=Open-source models, Rank=7.22025.08 | 56.5 | 77.2 | 81.5 | 57.7 | 68.2 |