GR-Dexter Technical Report

TL;DR

GR-Dexter introduces a comprehensive framework combining hardware, data collection, and training methods to enable vision-language conditioned manipulation on a bimanual dexterous robot, addressing challenges of high DoF and occlusions.

Contribution

It presents a novel hardware design, a teleoperation-based data collection system, and a training recipe leveraging large-scale datasets for generalist dexterous manipulation.

Findings

Achieves strong in-domain performance on real-world tasks.

Demonstrates robustness to unseen objects and instructions.

Enables long-horizon and generalizable pick-and-place tasks.

Abstract

Vision-language-action (VLA) models have enabled language-conditioned, long-horizon robot manipulation, but most existing systems are limited to grippers. Scaling VLA policies to bimanual robots with high degree-of-freedom (DoF) dexterous hands remains challenging due to the expanded action space, frequent hand-object occlusions, and the cost of collecting real-robot data. We present GR-Dexter, a holistic hardware-model-data framework for VLA-based generalist manipulation on a bimanual dexterous-hand robot. Our approach combines the design of a compact 21-DoF robotic hand, an intuitive bimanual teleoperation system for real-robot data collection, and a training recipe that leverages teleoperated robot trajectories together with large-scale vision-language and carefully curated cross-embodiment datasets. Across real-world evaluations spanning long-horizon everyday manipulation and generalizable pick-and-place, GR-Dexter achieves strong in-domain performance and improved robustness to unseen objects and unseen instructions. We hope GR-Dexter serves as a practical step toward generalist dexterous-hand robotic manipulation.