Physical AI Lab // Episode 3.5 // One-pager
GR00T fine-tuning in practice
Tuning an open humanoid model to your own robot
Three key points
- GR00T N1.7 is an open, commercially usable humanoid VLA: 3 billion parameters, a dual-system architecture (O), pre-trained on 20,854 hours of human egocentric video (O). A solid foundation.
- Fine-tuning starts with data: put it in the LeRobot format and write the modality config (wiring each dimension to a joint). Practise on a small example dataset first; which conversion scripts and examples the repo offers follows the repo's instructions.
- One command trains a little more, on a single GPU (GPU and memory requirements per the repo's instructions). Check it with an open-loop evaluation (predicted vs real actions), but good open-loop results don't mean a stable closed loop on a real robot.
Evidence: O official · R research · M media
| O | 3 billion | parameters: GR00T N1.7 is an open 3 billion parameter vision-language-action model (sees images, takes instructions, outputs continuous actions) |
| O | Commercial use | licence: NVIDIA describes GR00T N1.7 as open and commercially licensed |
| O | 20,854 hours | of human egocentric video used to pre-train GR00T N1.7 (EgoScale, over 20 task categories) |
| O | More than double | NVIDIA says: going from 1,000 to 20,000 hours of human video more than doubles average task completion (dexterity scaling law) |
| O | 22 DoF | hands: this dexterity law lets 22 degree-of-freedom dexterous hands do contact-rich fine tasks |
Further reading
- Isaac-GR00T repo (install, fine-tuning and evaluation commands)
https://github.com/NVIDIA/Isaac-GR00T - NVIDIA · GR00T N1.7 blog (3B parameters, EgoScale, dual system)
https://huggingface.co/blog/nvidia/gr00t-n1-7 - GR00T N1: a dual-system VLA foundation model (arXiv 2503.14734)
https://arxiv.org/abs/2503.14734