Tomato Grasping GR00T Model

Fine-tuned NVIDIA GR00T model for tomato grasping task.

Model Details

  • Base Model: nvidia/GR00T-N1.5-3B
  • Task: Tomato grasping and placement
  • Training Steps: 50,000
  • Final Loss: 0.037
  • Training Time: 4 hours 3 minutes

Hardware

  • Robot: SO101 Follower
  • Cameras: Wrist + Side view (Intel RealSense)
  • DOF: 6 (5 joints + gripper)

Dataset

  • Episodes: 40
  • Total Frames: 12,069
  • Task: "Grasp the tomato and place it in the container"
Downloads last month
3
Safetensors
Model size
3B params
Tensor type
F32
·
BF16
·
Video Preview
loading