skills.rest
bensonlee5
bensonlee5COMMUNITY
Explore all skills by this author →
GitHub

rldx1-ft-libero-nf4

Automate Franka Panda pick, place, open, and close actions using a vision-language-action policy.

Updated Jul 5, 2026
Search Tags:#vision-language-action#libero benchmark#rSkill#robotic manipulation#franka panda
One-click install
npx skills add https://github.com/bensonlee5/openral --skill rldx1-ft-libero-nf4
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rldx1-ft-libero-nf4
Source: https://github.com/bensonlee5/openral/tree/main/rskills/rldx1-ft-libero-nf4
Command: npx skills add https://github.com/bensonlee5/openral --skill rldx1-ft-libero-nf4

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires rldx, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates complex robotic manipulation tasks on a Franka Panda robot, streamlining operations and improving efficiency.

Core Features & Use Cases

  • Vision-Language-Action Policy: Executes pick, place, open, and close actions on objects like bowls, cups, drawers, and general objects.
  • LIBERO Benchmark Suite: Fine-tuned on the LIBERO benchmark task suite for robust performance.
  • Out-of-Process Sidecar: Runs via the rldx sidecar adapter for compatibility and performance optimization.
  • Quantization: NF4 quantization on the Qwen3-VL backbone for 8 GiB GPU compatibility.

Quick Start

Use the rldx1-ft-libero-nf4 skill to pick up a bowl on the Franka Panda robot.

Frequently Asked Questions about rldx1-ft-libero-nf4

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

FAQPage Schema
How do I automate pick and place tasks on a Franka Panda robot?▼

You can automate pick, place, open, and close actions on a Franka Panda robot by running a vision-language-action policy fine-tuned on the LIBERO benchmark task suite for tabletop and kitchen scenes.

Can I run vision-language-action robotic manipulation on an 8 GiB GPU?▼

Yes, this robotic manipulation approach uses NF4 quantization on the Qwen3-VL backbone to optimize vision-language-action policies specifically for compatibility with 8 GiB GPUs.

What is the LIBERO benchmark suite used for in robotic manipulation?▼

The LIBERO benchmark suite provides a set of tasks to fine-tune vision-language-action policies, ensuring robust performance for executing manipulation actions like pick and place on objects.

Do I need the rldx sidecar adapter for Franka Panda robotic manipulation?▼

Yes, the rldx out-of-process sidecar adapter is required to run the vision-language-action policy and execute manipulation tasks on the Franka Panda robot.

What objects can a vision-language-action policy interact with in tabletop scenes?▼

The vision-language-action policy can interact with objects like bowls, cups, and drawers to perform pick, place, open, and close actions within tabletop and kitchen scenes.

Related Skills

bensonlee5bensonlee5

topreward-qwen3vl-4b-nf4

Monitor robotic task progress with Qwen3-VL-4B VLM zero-shot reward model.

Community
Intermediate
bensonlee5bensonlee5

act-libero

Execute pick, place, open, and close actions via a Vision-Language-Action policy on LIBERO Franka-Panda.

Community
Intermediate
bensonlee5bensonlee5

pi05-libero-int8

Quantize the Vision-Language-Action policy for Franka Panda with int8.

Community
Advanced
bensonlee5bensonlee5

smolvla-vlabench

Automate Franka Panda pick, place, and grasp tasks with Vision-Language-Action policies.

Community
Advanced