RoboSynChallenge
EDEM-AI
A dexterous manipulation benchmark connecting synthesized skills, training data and evaluation.
What you can explore
Explore the released trajectories, ACT and Diffusion Policy training, checkpoints and versioned evaluation results.
Before you use it
Released simulation results do not establish real-world success. Preserve benchmark protocol, checkpoint revisions and environment versions when comparing methods.
Original contributors
RoboSynChallenge authors and EDEM-AI contributors
License / access: Code: Apache 2.0 · check external data, models and hardware terms separately
Keep exploring
ACT / Action Chunking with Transformers
An imitation-learning implementation that predicts chunks of robot actions from demonstrations.
Diffusion Policy
A visuomotor policy-learning framework that generates action sequences through diffusion.
DexMimicGen
Generated demonstrations and simulation environments for learning coordinated dexterous manipulation with two arms.
Editorial reference · checked 2026-09-22