---
tags:
- robotics
- world-model
- action-conditioned-video-generation
- robot-manipulation
- coachworld
pipeline_tag: robotics
---

# CoachWorld — Pretrained Robot World Model

[**Paper**](https://arxiv.org/abs/2609.39685) · [**Code**](https://github.com/RoboCoach-AI/CoachWorld) · [**Project Page & Videos**](https://robocoach-ai.github.io/)

**CoachWorld predicts future robot observations conditioned on visual history and robot actions.** It is the world-model component of **RoboCoach: World Models as Active Coaches for Compositional Robot Skills**.

This repository hosts the **CoachWorld pretrained checkpoint**, trained on heterogeneous single-arm and dual-arm robot data. 

## Overview

CoachWorld is an action-conditioned video world model built on an adapted Wan2.2 TI2V-5B backbone. It uses visual history and end-effector action conditioning to generate future observations, with camera and action conventions that support learning across robot datasets.

The model supports autoregressive rollout: generated observations become part of the context for subsequent predictions. In the full RoboCoach framework, skill policies interact with the world model, while a separate progress judge identifies subtask failures. These diagnoses guide targeted demonstration collection and policy updates.

## Model at a Glance

| Item | Description |
| --- | --- |
| Release | CoachWorld pretrain |
| Model type | Action-conditioned video world model |
| Backbone | Adapted Wan2.2 |
| Conditioning | Visual history and robot end-effector actions, using the prescribed camera/action representation |
| Output | Predicted future RGB observations |
| Rollout mode | Autoregressive video prediction |
| Training scope | Heterogeneous single-arm and dual-arm robot data |
| Implementation | Custom `coachworld` Python package |

## Getting Started

Download the weights from the [Files and versions](https://huggingface.co/JEdward/CoachWorld/tree/main) tab and use the [CoachWorld code repository](https://github.com/RoboCoach-AI/CoachWorld) for environment setup and model integration.

The implementation includes:

- Model conditioning, inference, and training components.
- Modified Wan2.2 model and VAE components.
- Video-latent data interfaces and action/camera conventions.
- Camera and robot geometry helpers.
- A distributed training entry point and example configuration.

**Input preparation matters.** Robot actions must follow the implementation’s coordinate-frame, normalization, and temporal-sampling conventions. Raw joint commands or pixel-space trajectories are not interchangeable with the model’s expected conditioning.

Use the CoachWorld implementation to load and run this checkpoint; this release does not claim compatibility with a generic Transformers or Diffusers loading pipeline.

## Citation

If you use CoachWorld in your research, please cite:

```bibtex
@article{liu2026robocoach,
  title={RoboCoach: World Models as Active Coaches for Compositional Robot Skills},
  author={Liu, Jiajun and Chen, Yifan and Liu, Yichao and Zhang, Jiayi
          and Chen, Ruoqu and Xie, Shaoxuan and Yao, Guocai
          and Xu, Mengdi and Cui, Sen and Zhang, Changshui},
  journal={arXiv preprint arXiv:2609.39685},
  year={2026},
  url={https://arxiv.org/abs/2609.39685}
}
```

## Acknowledgments

CoachWorld builds on Wan2.2 and other open-source components. See the code repository’s [third-party notices](https://github.com/RoboCoach-AI/CoachWorld/blob/main/THIRD_PARTY.md) for component attribution and applicable terms.

## Contact

For implementation questions and reproducibility issues, please open an issue in the [CoachWorld repository](https://github.com/RoboCoach-AI/CoachWorld/issues).