GR-Dexter: VLA for Bimanual Dexterous Robot Control

Organization: ByteDance Research

Overview

GR-Dexter is an integrated framework for enabling Vision-Language-Action (VLA) models to control bimanual dexterous robots. The system addresses challenges in scaling robotic manipulation beyond simple grippers to high-degree-of-freedom hands.

Key Hardware Components

ByteDexter V2 Hand:

  • 21 degrees of freedom (four per finger, five for thumb)
  • Dimensions: 219mm height × 108mm width
  • High-density piezoresistive sensor arrays for tactile feedback across fingertips

Teleoperation Interface (data collection):

  • Meta Quest VR for wrist tracking
  • Manus Metagloves for hand capture
  • Foot pedals for arm control
  • Enables coordination of two Franka arms for long-horizon manipulation tasks

Model Architecture

  • Parameters: 4 billion
  • Architecture: Mixture-of-Transformer (MoT)
  • Inputs: language instructions, visual observations, robot state
  • Outputs: action chunks controlling arm joints, end-effector poses, hand joints, fingertip positions

Training Strategy

Co-training on three data sources:

  1. Web-scale vision-language datasets
  2. Cross-embodiment robot data (Fourier ActionNet, OpenLoong, RoboMIND)
  3. Human VR trajectories (with temporal consistency filtering)

Performance Results

TaskConditionSuccess Rate
Long-Horizon TasksIn-distribution0.97
Long-Horizon TasksOut-of-distribution0.89
Long-Horizon Tasks (baseline)Out-of-distribution0.64
Pick-and-PlaceUnseen objects0.85
Pick-and-PlaceUnseen instructions0.83

Cross-embodiment data significantly improves generalization to novel scenarios while maintaining strong in-domain performance.