---
datasets: HuggingFaceVLA/libero
library_name: lerobot
license: apache-2.0
model_name: vla0_smol
pipeline_tag: robotics
tags:
- lerobot
- robotics
- vla0_smol
---

# Model Card for vla0-smol

Read about the model here: https://robot-learning-collective.github.io/vla-0-smol
This is VLA-0-Smol trained on the Libero dataset.

_Model type not recognized — please update this template._


This policy has been trained and pushed to the Hub using [LeRobot](https://github.com/huggingface/lerobot).
See the full documentation at [LeRobot Docs](https://huggingface.co/docs/lerobot/index).

---

## How to Get Started with the Model

For a complete walkthrough, see the [training guide](https://huggingface.co/docs/lerobot/il_robots#train-a-policy).

Support of the model is implemented in repo: https://github.com/Robot-Learning-Collective/lerobot-experiments


Below is the short version on how to train and run inference/eval:

### Train from scratch on PushT

```bash
lerobot-train --config_path=configs/vla0_smol.json
```

_Writes checkpoints to `outputs/train/<desired_policy_repo_id>/checkpoints/`._

### Evaluate the policy on Libero

```bash
lerobot-eval \
  --policy.path="Robot-Learning-Collective/VLA-0-Smol" \
  --policy.n_action_steps=0 \
  --policy.ensemble_size=8 \
  --env.type=libero \
  --env.task=libero_object \
  --eval.batch_size=2
```

Prefix the dataset repo with **eval\_** and supply `--policy.path` pointing to a local or hub checkpoint.

---

## Model Details

- **License:** apache-2.0