ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver
Paper • 2508.10333 • Published • 1
Quick Links
This repository contains the model introduced in ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver
Project page:https://zionchow.github.io/ReconVLA/
Code:https://github.com/Chowzy069/Reconvla
This model was pre-trained and then fine-tuned on CALVIN ABC->D.
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("zzyzyzy/ReconVLA", dtype="auto")