Gallery · LIBERO-Pro

Pick the alphabet soup and place it in the basket

libero_object_task, task 1 · seed 2. DeepSeek-V4.1-Flash, code mode: 2 decisions, 7 motions, solved.

2× · 29 s of motion
  1. 1 of 2 · step2.py
  2. 2 of 2 · step3.py
  1. set_gripper
  2. move_pose
  3. move_pose
  4. set_gripper
  5. move_pose
  6. move_pose
  7. move_pose
Decisions

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.