Foundry-VLA-1.7B-full
A 1.7B parameter vision-language-action model for bimanual robotic manipulation, part of the VLA Foundry collection. Trained on both simulated and real-world manipulation data.
Model Description
- Architecture: Foundry-VLM-1.3B vision-language backbone + (condition on last 4 layers) flow-matching diffusion action head (24 layers, 1024 hidden dim, 16 heads)
- Parameters: 1.7B (non-embedding)
- Action space: 20-dim relative actions (bimanual xyz + 6D rotation + gripper)
- Cameras: 4 views (2 scene + 2 wrist)
- Training data: 102M samples from simulated and real-world bimanual manipulation tasks (with resample)
- VLM backbone: Foundry-VLM-1.3B-200M
Evaluation Results
Success rates on 16 seen tasks and 3 unseen tasks (200 rollouts per task):
| Simulator | Seen (16 tasks) | Unseen (3 tasks) |
|---|---|---|
| CS | 54.0% | 4.7% |
| OSS | 34.1% | 7.5% |
Usage
git clone https://github.com/TRI-ML/vla_foundry.git
cd vla_foundry
pip install -e .
from vla_foundry.models.base_model import BaseModel
model = BaseModel.from_pretrained("TRI-ML/Foundry-VLA-1.7B-full")
Links
- Project page: tri-ml.github.io/vla_foundry
- Paper: VLA Foundry (arXiv 2604.19728)
- Code: github.com/TRI-ML/vla_foundry
- Collection: VLA Foundry collection