Robot Manipulation on RoboLab Moderate 120
23.3Vague Success RateCosmos3-Nano-Policy-DROID
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Cosmos3-Nano-Policy-DROIDCheckpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 23.3 | 35.4 | 40.3 | |
| DreamZeroCheckpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 19.5 | 30 | 26.7 | |
| Cosmos3-Nano (PT-init)Checkpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 19 | 28.7 | 29.5 | |
| π0.5Checkpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 17.9 | 31.5 | 31 | |
| π0-FASTCheckpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 12.8 | 13.3 | 12.6 | |
| paligemma-binningCheckpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 5.9 | 4.9 | 10.3 | |
| GR00T N1.6Checkpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 4.9 | 7.9 | 4.1 | |
| π0Checkpoint type=off-the-shelf, Training mode=fine-tuned on DROID dataset2026.06 | 3.8 | 3.6 | 2.1 |