Skip to main content

SuperDex Lab

SuperDex Lab connects simulation and policy development through a Gymnasium-style API for reinforcement learning.

Early Preview

SuperDex Lab is currently in early preview and will receive substantial improvements.

What's Included​

SuperDex Gym​

A Gymnasium-compatible reinforcement-learning framework built on the SuperDex Physics engine. Train agent controllers in high-fidelity simulation with a suite of manipulation and locomotion environments.

Start with the SuperDex Gym introduction, then install and set it up.

RLlib Training​

An off-the-shelf Ray/RLlib integration for RL training straight from the Gym environments. See Training with Ray/RLlib.

Benchmarking​

Benchmark control and simulation throughput and approximate memory usage across worker and environment configurations. See Benchmarking.