Place on Realman robot collected dataset Real-world In-Domain 1.0 (test)
87.5Pick Up SuccessVideoVLA
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VideoVLAPre-training=OXE dataset, Fine-tuning=collected dataset2025.12 | 87.5 | 25 | 56.3 | |
| CogACTPre-training=OXE dataset, Fine-tuning=collected dataset2025.12 | 54.2 | 16.7 | 35.5 | |
| π0Pre-training=OXE dataset, Fine-tuning=collected dataset2025.12 | 45.8 | 16.7 | 31.3 | |
| OpenVLAPre-training=OXE dataset, Fine-tuning=collected dataset2025.12 | 29.2 | 0 | 14.6 | |
| SpatialVLAPre-training=OXE dataset, Fine-tuning=collected dataset2025.12 | 20.8 | 0 | 10.4 |