---
base_model:
- nvidia/Cosmos-Predict2-2B-Video2World
license: other
license_name: nvidia-base-model-license
license_link: https://huggingface.co/nvidia/Cosmos-Predict2-2B-Video2World
pipeline_tag: robotics
tags:
- robotics
- world-action-model
- video-diffusion
- libero
---

# MetaWAM-LIBERO-Predict2-2B

[**Paper**](https://arxiv.org/abs/2611.14365) | [**Code**](https://github.com/showlab/MetaWAM) | *From Video Frames to Robot Meta-Frames: A Unified Latent Interface for World Action Models* (CoRL 2026)

MetaWAM is a World Action Model fine-tuned from [Cosmos-Predict2-2B-Video2World](https://huggingface.co/nvidia/Cosmos-Predict2-2B-Video2World). Robot signals (task type, proprioception, action chunks, future proprioception and values) are packed into fixed coefficient slots of disjoint channel bands and lifted into compact *meta-frames* with a 2D inverse DCT. Slot-aware masks then train policy, world-model and value prediction in a single denoising model.

This checkpoint is trained on **LIBERO** (LIBERO-Spatial, -Object, -Goal and -10, trained jointly).

## Files

| File | Description |
|---|---|
| `MetaWAM-LIBERO-Predict2-2B.pt` | Model weights (bf16, consolidated) |
| `libero_dataset_statistics.json` | Action / proprio normalization statistics |
| `libero_t5_embeddings.pkl` | Precomputed T5 embeddings of the task instructions |
| `SHA256SUMS` | Checksums |

The statistics and T5 embedding files are identical to the ones released with Cosmos Policy.

## Usage

Use the inference config `metawam_libero__inference` from the [MetaWAM repository](https://github.com/showlab/MetaWAM):

```bash
hf download Ethanliang99/MetaWAM-LIBERO-Predict2-2B --local-dir checkpoints/MetaWAM-LIBERO-Predict2-2B
bash scripts/test_libero.sh     # 4 suites, seeds 195-197
```

See [LIBERO.md](https://github.com/showlab/MetaWAM/blob/main/LIBERO.md) for details.

## License

These weights are fine-tuned from `nvidia/Cosmos-Predict2-2B-Video2World`; use is subject to the license of the base model.

## Citation

```bibtex
@article{liang2026metawam,
  title   = {From Video Frames to Robot Meta-Frames: A Unified Latent Interface for World Action Models},
  author  = {Liang, Guoqiang and Lin, Yiqi and Chen, Yanzhe and Ma, Kevin Yuchen and Shou, Mike Zheng},
  journal = {arXiv preprint arXiv:2611.14365},
  year    = {2026}
}
```