New Report

The Humanoid Robot Supply Chain

Supplier Strategy and Market Positioning 2026–2027

Get the Report

New! 2026 Humanoid
Robot Market Report

198 pages of exclusive insight from global robotics experts — uncover funding trends, technology challenges, leading manufacturers, supply chain shifts, and surveys and forecasts on future humanoid applications.

Aaron Saunders
Featuring insights from Aaron Saunders, Former CTO of Boston Dynamics,
now Google DeepMind
Get the Report
New Report

Humanoid Foundation Models

The brains are being rebuilt

Get the Report

GR-Dexter

$ 0

ByteDance logo humanoid guide

GR-Dexter is ByteDance Seed’s robot foundation model for dexterous hands. It controls two five-fingered hands to handle complex everyday tasks and copes well with new objects and instructions.

Available on backorder

Capabilities
5
Humanoid.Guide
Brain Score
=Brain score5
Brain score filledBrain score filledBrain score filledBrain score filledBrain score filledBrain score unfilledBrain score unfilledBrain score unfilledBrain score unfilledBrain score unfilled


Specifications and details:

Nationality China
Website https://seed.bytedance.com/en/
Model type Foundation Model
Manufacturer ByteDance Seed (ByteDance)
Release date December 2025

Description

GR-Dexter is a robot foundation model from ByteDance Seed, the AI research team of TikTok’s parent company. The team introduced it in December 2025 to tackle one of robotics’ hardest problems: dexterous hands. Most robot models control simple two-finger grippers. GR-Dexter, however, controls two human-like hands with five fingers each. It’s a vision-language-action model, so it turns camera images and spoken instructions into precise hand movements. In addition, it builds on ByteDance’s earlier GR-3 model.

ByteDance paired the model with its own ByteDexter robotic hand, which has touch sensors in the fingertips. The team trained GR-Dexter on several kinds of data. These include remote-controlled robot demonstrations, data from other robots and videos of people using their hands. As a result, the model copes well with objects and instructions it has never seen. For example, it tidied makeup items, used a vacuum and served bread. Moreover, it stayed reliable when the scene changed in unexpected ways.

New Report

The Humanoid Robot Supply Chain

Supplier Strategy and Market Positioning 2026–2027

Get the Report

New! 2026 Humanoid
Robot Market Report

198 pages of exclusive insight from global robotics experts — uncover funding trends, technology challenges, leading manufacturers, supply chain shifts, and surveys and forecasts on future humanoid applications.

Aaron Saunders
Featuring insights from Aaron Saunders, Former CTO of Boston Dynamics,
now Google DeepMind
Get the Report
New Report

Humanoid Foundation Models

The brains are being rebuilt

Get the Report

Contact Humanoid.guide

Website: https://byte-dexter.github.io/gr-dexter