PyTorch code and models for VJEPA2 self-supervised learning from video.
Repository Index
105,102 visible repositories in the index
1–8 of 8 · Most stars
Simulation Platform from AgiBot
robomimic: A Modular Framework for Robot Learning from Demonstration
Base code for the Dingo quadruped; modified from Stanford Pupper and Notspot repositories. Includes integration with ROS Noetic and a simulation of the Dingo
Independent reproduction of SuperMimic (arXiv:2610.28480) on mjlab/MuJoCo Warp, written from the paper before the official release. For the authors' code see HybridRobotics/supermimic. Kept as a record: experiment log, full BONES-SEED results, checkpoint.
- humanoid
- mjlab
- motion-tracking
- mujoco
HALO: Perceptive Humanoid Motion from Simulation
The first exam for code agents in the physical world: frozen robot tasks from many benchmark suites, each answered with a program and verified in a separate sandbox. Runs on Harbor.
- ai-agents
- benchmark
- coding-agents
- embodied-ai
Whole-body motion tracking for the Unitree G1 in under 3 minutes on average on one GPU: return 0.82 with 0/40 clips failed, held from 2.74 min (mean of 15 seeds, range 2.21-3.61, fine checkpoints), 11.87 it/s (world model + Muon + overlapped CUDA streams, mjlab)
- humanoid
- robot-learning
- supermimic
- world-model