Introducing SimReadyGen

From a text prompt to physically accurate SimReady assets.

Physical AI runs on simulation. Robot training now demands physically accurate environments at a scale and speed that traditional asset creation can't match. Today we're launching SimReadyGen, Lightwheel's agentic simulation-generation engine. Built on OpenUSD and integrated with NVIDIA Omniverse Libraries, SimReadyGen generates structured, simulation-ready assets from a text prompt.

Measured Physics, Generated at Scale

Generated assets are only useful for robot training if their physics hold up in the real world. Behind SimReadyGen is SimReady Foundry, our measured-physics pipeline. It begins in our Physics Measurement Factory, where we measure how real objects actually behave: contact, friction, and dynamics. Those measurements become ground-truth physical parameters for simulation.

The same measurements also drive the development of our physics solver, so simulated behavior matches how objects behave in the real world. The result is a large, growing library of physically measured SimReady assets. SimReadyGen generates on top of Omniverse Libraries, drawing on this foundation of measured physics. Every asset it produces carries measured physics, not estimates, at generation speed.

Built for OpenUSD Workflows

We integrated NVIDIA Omniverse Content Agents to support automated 3D content workflows for USD files, including material assignment, physics property classification, texture generation, and content validation. SimReadyGen is built on OpenUSD, so every asset and environment it creates can move through robotics and simulation workflows, including Isaac Sim or Isaac Lab.

Lightwheel's Continuous Learning System for Physical AI

SimReadyGen is one part of a larger loop. Generated assets and environments feed RoboFinals, our Industrial-Grade evaluation platform, where policies are tested against measured physics before they touch hardware. Evaluation results flow into RoboStack, our end-to-end deployment pipeline, and real-world performance data returns to refine the next round of simulation. Generate, evaluate, deploy, and learn: this is the continuous learning system we're building for Physical AI, and SimReadyGen is where each cycle begins.

Get started with SimReadyGen →