train-policy

Name: train-policy
Availability: InStock
Author: Nebius

Coordinates robot-policy training lifecycle across LeRobot, Isaac Lab, and SONIC via YAML specifications and Nebius GPU/S3 infrastructure.

17|8|Updated Apr 7, 2026

Search Tags:#workflow #machine learning #gpu #robotics #nebius #policy training

Ready to run

npx skills add https://github.com/nebius/nebius-physical-ai --skill train-policy

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: train-policy
Source: https://github.com/nebius/nebius-physical-ai/tree/main/skills/workflows/train-policy
Command: npx skills add https://github.com/nebius/nebius-physical-ai --skill train-policy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill streamlines the complex process of configuring, launching, and monitoring robot policy training workflows across diverse frameworks like LeRobot, Isaac Lab, and SONIC.

Core Features & Use Cases

Unified Workflow Management: Provides a single interface to manage training, evaluation, and export tasks for various robot policy families.
Infrastructure Integration: Seamlessly connects training jobs to Nebius GPU clusters and S3 storage for artifact management.
Use Case: A robotics team can use this skill to transition from local simulation testing to large-scale policy training on H100 clusters by simply selecting the appropriate workflow YAML and GPU target.

Quick Start

Use the train-policy skill to initiate a training run for your Isaac Lab configuration by specifying the target GPU and the corresponding workflow YAML file.

Frequently Asked Questions about train-policy

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

FAQPage Schema

How do I orchestrate robot policy training across multiple simulation frameworks?▼

Robot policy training is coordinated end-to-end across frameworks like LeRobot, Isaac Lab, and SONIC using declarative YAML specifications. This unified workflow approach manages planning, execution, and evaluation while integrating with Nebius infrastructure for GPU allocation and S3-based artifact management.

How do I run large-scale robot policy training on Nebius H100 GPU clusters?▼

Large-scale robot policy training on Nebius H100 clusters is executed by specifying a target GPU and a workflow YAML file. This leverages Nebius infrastructure for scalable GPU allocation and S3 storage to manage artifacts, ensuring reproducible training workflows.

Does this policy training workflow support LeRobot and Isaac Lab environments?▼

Yes, the policy training workflow supports diverse environments including LeRobot, Isaac Lab, and SONIC. It coordinates training, evaluation, and export tasks across these frameworks by leveraging NPA workbench tools and declarative YAML specifications.

What's the best way to manage reproducible machine learning artifacts during GPU training?▼

Managing reproducible machine learning artifacts is best achieved by integrating training workflows with S3-based artifact management. This approach ensures reproducible and scalable training runs by storing outputs alongside declarative YAML configurations and Nebius GPU infrastructure allocations.

Do I need Nebius infrastructure to use this robot policy training skill?▼

Yes, Nebius infrastructure integration is required for GPU allocation and S3-based artifact management to ensure reproducible and scalable training workflows. The skill coordinates the end-to-end lifecycle across multiple frameworks by relying on this infrastructure.

How do I evaluate and export robot policies after training is complete?▼

Evaluating and exporting robot policies is handled within the unified workflow management interface. After training execution, the skill coordinates evaluation and export tasks for various robot policy families using declarative YAML specifications.