GR00T-H

Model

A vision-language-action model for surgical robotics developed by NVIDIA. It is a post-trained variant of the NVIDIA Isaac GR00T N1.6 foundation, adapted for surgical robots using the Open-H embodiment dataset.

Details

Updated:8/20/2026

open sourcetrue

release date2026-08-19

github urlhttps://github.com/NVIDIA-Medtech/GR00T-H

model familyVLA

huggingface urlhttps://huggingface.co/nvidia/GR00T-H

Relationships

NVIDIA

Company

DEVELOPED_BY

Sources

GR00T-H model card on Hugging Face

website

Visit

GR00T-H official repository