Metaworld

Robotics multi-task and meta-reinforcement learning benchmark (Meta-World). Standardized continuous-control benchmark built on Gymnasium with MT1, MT10, MT50 multi-task suites and ML1, ML10, ML45 meta-learning suites. Supports synchronous and asynchronous vector execution. Use for robotic manipulation benchmarking, multi-task RL, meta-RL adaptation, and evaluating generalization across tasks and goals.

mkurman Updated 327 repo stars
Picture from the README

File contents

mkurman/zorai commit 5e0f2b44cc

Frequently asked questions

npx get-skill add mkurman/metaworld