add model README

README.md

@@ -1,3 +1,50 @@
 ---
 license: mit
 ---
+<div align="center">
+<h1 align="center"> 🔬 LabVLA </h1>
+<h3 align="center"> Grounding Vision–Language–Action Models in Scientific Laboratories </h3>
+
+<p align="center">
+  <a href="https://zjunlp.github.io/LabVLA/">🔥Project Page</a> •
+  <a href="https://github.com/zjunlp/LabVLA">💻GitHub Repo</a> •
+  <a href="https://huggingface.co/zjunlp/LabVLA">🤗Model</a>
+</p>
+</div>
+
+---
+
+## Model Description
+
+**LabVLA** is the first vision–language–action (VLA) model designed for scientific laboratory environments. It combines a **Qwen3-VL-4B** vision–language backbone with a **DiT flow-matching action expert**, trained with the π0.5 recipe to enable real-time robot control in lab settings.
+
+The checkpoint is **self-contained** — loading requires no external base model. The bundled `vlm_config.json` allows the backbone to be rebuilt weightlessly, and `model.safetensors` provides all weights.
+
+## How to Use
+
+### Download
+
+```bash
+huggingface-cli download zjunlp/LabVLA --local-dir LabVLA
+```
+
+### Deployment
+
+Serve the model over the OpenPI msgpack WebSocket protocol:
+
+```bash
+git clone https://github.com/zjunlp/LabVLA.git
+cd LabVLA
+bash deployment/deploy.sh
+```
+
+### Load in Python
+
+```python
+import sys; sys.path.insert(0, "src")
+from policies.LabVLA.modeling_labvla import LabVLAPolicy
+
+policy = LabVLAPolicy.from_pretrained("zjunlp/LabVLA")
+```
+
+For training, data preparation, and more details, please refer to our [GitHub repository](https://github.com/zjunlp/LabVLA).