Diffusers
Safetensors
WorldDirector / README.md
nielsr's picture
nielsr HF Staff
Add model card, pipeline tag, and links to paper/code
52788eb verified
|
Raw
History Blame
2.07 kB
---
license: cc-by-nc-sa-4.0
pipeline_tag: image-to-video
---
# WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory
[πŸ“„ Paper](https://huggingface.co/papers/2607.02517) | [🌐 Project Page](https://worlddirector.github.io/) | [πŸ™ GitHub Repository](https://github.com/pPetrichor/WorldDirector)
**WorldDirector** is a highly controllable video world model framework designed for persistent dynamic object memory and unrestricted viewpoint exploration.
Unlike existing world models that entangle physical dynamics with pixel rendering and rely on continuous visual observation to sustain motion, this framework explicitly decouples semantic motion orchestration from visual generation. By leveraging an LLM to coordinate 3D trajectories with camera movements and subsequently employing these orchestrated trajectories as control signals for video generation, this approach ensures strict physical logic and appearance stability, successfully preserving the exact visual identities of dynamic entities even when they re-enter the scene after prolonged periods out of view.
## Setup & Inference
For instructions on environment setup, preparing latents/conditions, and running the causal video generation pipeline, please refer to the [WorldDirector GitHub repository](https://github.com/pPetrichor/WorldDirector).
### Model Download via CLI
You can download the model weights using `huggingface-cli` as follows:
```shell
pip install "huggingface_hub[cli]"
huggingface-cli download hlwang06/WorldDirector --local-dir ./WorldDirector-14B
```
## Citation
If you find this work useful, please consider citing our paper:
```bibtex
@article{wang2026worlddirector,
title={WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory},
author={Hanlin Wang and Hao Ouyang and Qiuyu Wang and Wen Wang and Qingyan Bai and Ka Leong Cheng and Yue Yu and Yixuan Li and Yihao Meng and Zichen Liu and Yanhong Zeng and Yujun Shen and Qifeng Chen},
journal={arXiv preprint arXiv:2607.02517},
year={2026}
}
```