Open the middle layer of the drawer
libero_goal_swap, task 0 · seed 3. DeepSeek-V4.1-Flash, code mode: 7 decisions, 13 motions, solved.
2× · 37 s of motion
- 1 of 7 · view1.py
- 2 of 7 · view2.py
- 3 of 7 · step_pull.py
- 4 of 7 · step_grasp2.py
- 5 of 7 · step3.py
- 6 of 7 · step4.py
- 7 of 7 · step5.py
- move_to
- move_pose
- move_pose
- move_pose
- move_pose
- move_pose
- move_pose
- move_to
- move_pose
- set_gripper
- move_to
- move_to
- move_to
Decisions
libero_goal_swap-t00 libero_goal_swap · task 0 · seed 3: Code mode · 7 decisions, solved
The recording is the robot's own cameras at 2×; the time between motions, while the model thinks, is not filmed. The 3D replay is the camera's picture in depth after each decision, with the paths it planned, to turn and step through.