Robotic Manipulation on Real-world ALOHA OOD
88.5Color FidelityQwen-VLA-aloha
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Qwen-VLA-alohaPre-training strategy=fine-tuned from pretrained Qwen-VLA-Base2026.05 | 88.5 | 76.9 | 53.8 | 80.8 | 84.6 | 76.9 | |
| π0.52026.05 | 57.7 | 61.5 | 19.2 | 26.9 | 42.3 | 41.5 | |
| GR00T N1.62026.05 | 46.2 | 38.5 | 3.8 | 19.2 | 19.2 | 25.4 | |
| Qwen-VLA-alohaPre-training strategy=trained from scratch2026.05 | 42.3 | 30.8 | 34.6 | 30.8 | 42.3 | 36.2 |