
This is Eden: a robot learning framework for agents that enables them to build tasks faster in Isaac Lab. Eden can perform door opening 2.1x faster than IsaacLab. It can also handle bin picking 4x faster. As they explain SONIC, BeyondMimic and 21 published policies run as shipped.
Two years ago, we released Genesis, a universal physics engine for robotics.
Today, I’m excited to announce 🍎 Eden — a robot learning framework built from the ground up for agents, unleashing full potential of Genesis.
One task definition. From simulation to real robots. Sim-to-sim and sim-to-real just got easier.
Coding agents build tasks 1.6–4× faster than in Isaac Lab.
🧵 1/11
— Kashu Yamazaki (@kashu_yamazaki) November 14, 2026
Eden supports humanoids, quadrupeds, arms and dexterous hands. It lets you load “human-motion datasets, retarget them to robots, estimate motion from video, teleoperate with GELLO, Meta Quest, PICO or the Wuji Glove, and record or script demonstrations into HDF5 and LeRobot datasets.”
[HT]












































