Update README.md
Browse files
README.md
CHANGED
|
@@ -18,8 +18,6 @@ license: mit
|
|
| 18 |
|
| 19 |
**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.
|
| 20 |
|
| 21 |
-
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.
|
| 22 |
-
|
| 23 |
## How to Use
|
| 24 |
|
| 25 |
### Download
|
|
|
|
| 18 |
|
| 19 |
**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.
|
| 20 |
|
|
|
|
|
|
|
| 21 |
## How to Use
|
| 22 |
|
| 23 |
### Download
|