Robot learning
Policies trained from data instead of hand-written controllers.
- SmolVLAOpen source
A compact, efficient VLA trained on community LeRobot datasets.
- openpiOpen source
Open weights and code for Physical Intelligence's π0 robot foundation models.
- NVIDIA Isaac GR00T N1Open source
Open foundation model for generalist humanoid robots.
- RT-1Open source
Robotics Transformer 1 and its 130k-episode dataset of real-world tasks.
