---
license: apache-2.0
pipeline_tag: robotics
library_name: pytorch
tags:
  - fast-vla
  - fastvla
  - vision-language-action
  - vla
  - robotics
  - robot-manipulation
  - robotwin
  - libero
base_model:
  - facebook/dinov2-base
---

# Fast-VLA (FastVLA) checkpoints

[Paper (arXiv:2610.21389)](https://arxiv.org/abs/2610.21389) · [Project page](https://garfield-kh.github.io/FastVLA) · [Code](https://github.com/Garfield-kh/FastVLA)

**Fast-VLA** calls a frozen 4B vision-language model (DM0.5) **once per episode** and keeps its
layer-wise key/value cache as a fixed task memory. At every action chunk only a lightweight
online actor runs: it reads the current views and robot state, queries the task memory, and
predicts the next chunk of actions.

This repository holds the online actors (DINOv2-based encoders + action decoder). The frozen
DM0.5 VLM is downloaded from the Hugging Face Hub on first use.

| Folder | Benchmark | Trained on | Parameters | Success rate (%) |
|---|---|---|---|---|
| `fastvla-robotwin/` | RoboTwin 2.0 | 27,500 demos (50 clean + 500 randomized per task) | 272M | 92.46 Clean / 92.40 Randomized |
| `fastvla-libero/` | LIBERO | 2,000 demos (Spatial / Object / Goal / Long) | 185M | 95.95 LIBERO / 79.11 LIBERO-Plus |

Each folder contains `model.safetensors`, `config.json`, `norm_stats.json` and `rope_tables.pt`.

## Usage

```bash
git clone https://github.com/Garfield-kh/FastVLA && cd FastVLA
hf download kehong/FastVLA-weights --local-dir checkpoints   # or --include "fastvla-libero/*"
```

```python
from fastvla.policy import FastVLAPolicy

policy = FastVLAPolicy("checkpoints/fastvla-robotwin")
policy.reset()                                           # start of an episode
actions = policy.act([head_rgb, left_wrist_rgb, right_wrist_rgb], joint_targets,
                     "Place the cans in the plastic box.")   # (50, 14)
```

Installation, evaluation and training instructions are in the
[GitHub repository](https://github.com/Garfield-kh/FastVLA).

## Citation

```bibtex
@article{gong2026fastvla,
  title   = {Fast-VLA: You Only Think Once for Robot Manipulation},
  author  = {Gong, Kehong and Hou, Guanli and Zheng, Xiaoxu and Wang, Qi and Xu, Mingxi and
             Zhai, Anbang and He, Xiaoyu and Zhang, Mingyuan and Mi, Michael Bi},
  journal = {arXiv preprint arXiv:2610.21389},
  year    = {2026}
}
```

## License

The checkpoints are released under Apache-2.0. DM0.5, which the policy uses frozen, is
distributed separately under its own license.