rldx1-ft-simpler-widowx-nf4

Community

Automate robot manipulation tasks with RLDX-1 fine-tuned on SIMPLER WidowX benchmark.

Authorbensonlee5
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill addresses the challenge of automating robot manipulation tasks, particularly for robots with WidowX embodiments, by providing a fine-tuned Vision-Language-Action (VLA) policy.

Core Features & Use Cases

  • Vision-Language-Action Policy: Employs RLDX-1, a VLA policy, for precise robot manipulation.
  • Benchmarked on SIMPLER WidowX: Ensures effective real-to-sim correlation for the Bridge / WidowX evaluation protocol.
  • Non-commercial License: Designed for non-commercial use cases.
  • Use Case: This Skill can be used in scenarios where a robot needs to perform actions like picking up objects and placing them on a surface, ideal for tasks in simulated environments or in real-world settings with WidowX robots.

Quick Start

Load and execute the skill using the following command:

openral rskill run OpenRAL/rskill-rldx1-ft-simpler-widowx-nf4 --action pick --object object --scene tabletop

Dependency Matrix

Required Modules

openralpytorchtransformers

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: rldx1-ft-simpler-widowx-nf4
Download link: https://github.com/bensonlee5/openral/archive/main.zip#rldx1-ft-simpler-widowx-nf4

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.