GR00T N1
NVIDIA's open foundation model for humanoid robots
NVIDIA built a brain for humanoid robots — and then made it free to download. It handles both thinking (what should I do?) and movement (how do I do it?) in one package, designed specifically for robots shaped like humans.
NVIDIA provides excellent tooling via Isaac, making setup more straightforward than typical research models. Still requires a capable GPU and familiarity with robotics simulators.
Open foundation model for generalized humanoid robot reasoning and skills. Dual-system architecture combining vision-language processing with a diffusion transformer for real-time motor actions.
Real robot trajectories, human videos, synthetic simulation data
Control a full humanoid body from a language command
Bimanual assembly of complex parts
Mobile manipulation in unstructured environments
Serious GPU and ML engineering required
RELATED DISCUSSIONS
Community →Seeded and planned prompts — not live forum activity yet.
- Community signalIntermediate / AdvancedHow should EAR track open robotics models?
Meta-discussion on ontology, benchmarks, and graph links for open-weight VLAs and manipulation models.
- Community signalIntermediate / AdvancedOpenVLA in the field — what breaks first?
Open Models room prompt linked to openvla-7b — inference, fine-tuning, and sim-to-real gaps. Seeded prompt only.