Views
No views yet
1from gr00t.policy.policy import Gr00tPolicy
2
3policy = Gr00tPolicy.from_pretrained(
4 "TJ-chen/GR00T-N1.6-LIBERO-Long",
5 device="cuda:0"
6)
7
8# Run inference
9action, info = policy.get_action(observation)train_config.json for full training details.1@inproceedings{gr00tn1_2025,
2 title={GR00T N1: An Open Foundation Model for Generalist Humanoid Robots},
3 author={NVIDIA},
4 booktitle={ArXiv Preprint},
5 year={2025},
6}