README.md

---
pipeline_tag: image-to-video
library_name: diffusers
language:
  - en
tags:
  - world-model
  - video-generation
  - image-to-video
---

<div align="center">

<img src="https://mbzuai-oryx.github.io/WorldGuide/assets/worldguide-logo.png" alt="WorldGuide" width="120"/>

# WorldGuide: Goal-Directed Video World Model for Procedural Task Execution

[Ankan Deria](https://ankan8145.github.io), [Komal Kumar](https://komalkumar.org), [Hisham Cholakkal](https://hishamcholakkal.com), [Fahad Shahbaz Khan](https://sites.google.com/view/fahadkhans), [Salman Khan](https://salman-h-khan.github.io)

**Mohamed bin Zayed University of Artificial Intelligence**

[![Project Page](https://img.shields.io/badge/Project-Page-2ea44f?logo=googlechrome&logoColor=white)](https://mbzuai-oryx.github.io/WorldGuide/)
[![GitHub](https://img.shields.io/badge/GitHub-Code-181717?logo=github&logoColor=white)](https://github.com/mbzuai-oryx/WorldGuide)

</div>

Official checkpoints for **WorldGuide**. Method details, results and training code are on the [project page](https://mbzuai-oryx.github.io/WorldGuide/) and on [GitHub](https://github.com/mbzuai-oryx/WorldGuide).

## Checkpoints

```text
WorldGuide-Ckpt/
├── transformer/          # WorldGuide Video DiT        → --action_ckpt
└── text_encoder/llm/     # WorldGuide ContextPlanner   → --planner_model_path
```

## Usage

```bash
git clone https://github.com/mbzuai-oryx/WorldGuide.git
cd WorldGuide
pip install -r requirements.txt

# Download this repository and all other required components into ./ckpts
python download_models.py --hf_token <your_hf_token>

# Run closed-loop inference
bash scripts/inference/run_v5_qwen_closed_loop_memory.sh
```

## Citation

```bibtex
@misc{worldguide,
  title  = {WorldGuide: Goal-Directed Video World Model for Procedural Task Execution},
  author = {Deria, Ankan and Kumar, Komal and Cholakkal, Hisham and Khan, Fahad Shahbaz and Khan, Salman},
  year   = {2026}
}
```