Task Planning on EgoPlan-Benchmark1 (val)
47.4AccuracyPhysBrain-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| PhysBrain-8BMethod Category=Embodied Brain2025.12 | 47.4 | |
| RoboBrain2.5-8BMethod Category=Embodied Brain2025.12 | 45.9 | |
| Qwen-3-VL-8BMethod Category=General VLM2025.12 | 44.3 | |
| PhysBrain-4BMethod Category=Embodied Brain2025.12 | 43.9 | |
| Qwen-3-VL-4BMethod Category=General VLM2025.12 | 42.2 | |
| VST-RL-7BMethod Category=Embodied Brain2025.12 | 40.8 | |
| GPT-4oMethod Category=General VLM2025.12 | 39.5 | |
| RoboBrain2.0-7BMethod Category=Embodied Brain2025.12 | 38.6 | |
| MiniGPT-4-7BMethod Category=General VLM2025.12 | 28.1 | |
| LLaVA-1.5-7BMethod Category=General VLM2025.12 | 27.8 | |
| LLaMA-3.2-11BMethod Category=General VLM2025.12 | 24.3 |