isaac-lab

Orchestrate Isaac Lab reinforcement learning simulations on Nebius GPU clusters.

17|8|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/nebius/nebius-physical-ai --skill isaac-lab-nebius
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: isaac-lab
Source: https://github.com/nebius/nebius-physical-ai/tree/main/skills/tools/isaac-lab
Command: npx skills add https://github.com/nebius/nebius-physical-ai --skill isaac-lab-nebius

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires boto3, and includes scripts (resource) components.

What problem does it solve?

This skill streamlines the complex setup and execution of Reinforcement Learning (RL) simulation workflows within the Isaac Lab framework, ensuring hardware compatibility and operational consistency.

Core Features & Use Cases

  • Simulation Orchestration: Manage training, evaluation, and deployment of RL agents on specialized GPU clusters.
  • Hardware-Aware Execution: Automatically enforces constraints for RT-core enabled hardware like L40S or RTX Pro 6000.
  • Sim2Real Integration: Supports pluggable backends for held-out rollout evaluation, ensuring consistent reporting across different simulation engines.

Quick Start

Use the isaac-lab skill to deploy a new training job on the cluster by running the workbench deploy command.

Frequently Asked Questions about isaac-lab

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy and train reinforcement learning models in Isaac Lab on GPU clusters?

This skill orchestrates reinforcement learning simulation workflows in Isaac Lab on Nebius GPU infrastructure, managing training, evaluation, and deployment cycles for robotics policies via the workbench deploy command.

Does Isaac Lab reinforcement learning training require specific RT-core GPU hardware?

Yes, Isaac Lab training enforces hardware-aware execution constraints, automatically requiring RT-core enabled GPUs like L40S or RTX Pro 6000 to run reinforcement learning simulation workloads.

Can I use SkyPilot for cluster management when running Isaac Lab simulations?

Yes, Isaac Lab simulations require SkyPilot integration for cluster management, enabling you to orchestrate GPU computing resources and manage reinforcement learning training and evaluation cycles across the cluster.

How does Sim2Real integration work for held-out rollout evaluation in reinforcement learning?

Sim2Real integration supports pluggable backends for held-out rollout evaluation, ensuring consistent reporting of reinforcement learning policy performance across different simulation engines during deployment.

Can I override the Isaac Lab framework to use a custom fork for specialized research environments?

Yes, Isaac Lab supports custom fork overrides for specialized research environments, allowing you to tailor reinforcement learning simulation workflows and robotics policy training to your specific research requirements.

What is the best way to manage robotics policy evaluation and deployment cycles in Isaac Lab?

The best way to manage robotics policy evaluation and deployment cycles in Isaac Lab is using this skill to orchestrate reinforcement learning simulation workflows on Nebius GPU clusters while enforcing hardware-specific constraints.