Spatial Reasoning on SAT (Val, Real, MV metrics)
87.7Val Metric ScoreSAT-Dynamic-13B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SAT-Dynamic-13BModel Category=Open-source models, Rank=6.62025.08 | 87.7 | 54.9 | 44.4 | |
| FSD-13BModel Category=Open-source models, Rank=4.62025.08 | 73.2 | 63.3 | 46.6 | |
| Embodied-R1 w/o CSModel Category=Open-source models, Rank=3.42025.08 | 70 | 73.9 | 47.4 | |
| Embodied-R1Model Category=Open-source models, Rank=2.12025.08 | 70 | 76.3 | 51.1 | |
| ASMv2-13BModel Category=Open-source models, Rank=7.22025.08 | 63.9 | 46.7 | 44.4 | |
| Embodied-SFTModel Category=Open-source models, Rank=3.72025.08 | 59.3 | 65.5 | 50.4 | |
| RoboPoint-13BModel Category=Open-source models, Rank=7.22025.08 | 53.3 | 46.6 | 44.4 | |
| LLaVA-1.5-13BModel Category=Open-source models, Rank=9.42025.08 | 51.4 | 41.6 | 41.4 | |
| GPT4oModel Category=Closed-source models2025.08 | 49.4 | 57.5 | 60.2 | |
| Qwen2.5VL-3BModel Category=Open-source models, Rank=5.62025.08 | 48.7 | 45.1 | 44.4 | |
| RoboBrain-7BModel Category=Open-source models, Rank=4.42025.08 | 45.3 | 52.2 | 55.6 | |
| GPT4VModel Category=Closed-source models2025.08 | 44.8 | 50.7 | 55.6 |