MetaWorld VLA with OpenAI CLIP ViT

Overview

This repository provides a lightweight Vision-Language-Action baseline for MetaWorld robot-arm tasks utilizing a pretrained CLIP-ViT vision transformer. It is designed for developers and researchers working on robotics and AI integration in vision-language tasks.