ORIGINAL_README.md

<div align="center">

<div align="center">
  
<h1>Reward Forcing: <br> Efficient Streaming Video Generation with <br> Rewarded Distribution Matching Distillation</h1>

<div>
  <a href="#" target="_blank">Yunhong Lu</a><sup>1,2</sup>,
  <a href="https://zengyh1900.github.io/" target="_blank">Yanhong Zeng</a><sup>2</sup>,
  <a href="#" target="_blank">Haobo Li</a><sup>2,4</sup>,
  <a href="https://ken-ouyang.github.io/" target="_blank">Hao Ouyang</a><sup>2</sup>,
  <a href="https://github.com/qiuyu96" target="_blank">Qiuyu Wang</a><sup>2</sup>,
  <a href="https://felixcheng97.github.io/" target="_blank">Ka Leong Cheng</a><sup>2</sup>,
  <br>
  <a href="#" target="_blank">Jiapeng Zhu</a><sup>2</sup>,
  <a href="#" target="_blank">Hengyuan Cao</a><sup>1</sup>,
  <a href="https://zhipengzhang.cn/" target="_blank">Zhipeng Zhang</a><sup>5</sup>,
  <a href="https://openreview.net/profile?id=%7EXing_Zhu2" target="_blank">Xing Zhu</a><sup>2</sup>,
  <a href="https://shenyujun.github.io/" target="_blank">Yujun Shen</a><sup>2</sup>,
  <a href="#" target="_blank">Min Zhang</a><sup>1,3</sup>
</div>

<br>

<div>
  <sup>1</sup>ZJU, 
  <sup>2</sup>Ant Group, 
  <sup>3</sup>SIAS-ZJU, 
  <sup>4</sup>HUST, 
  <sup>5</sup>SJTU
</div>

</div>

<br>

[![Paper](https://img.shields.io/badge/Paper-arXiv-red)](https://arxiv.org/abs/2512.04678)
[![Project Page](https://img.shields.io/badge/Project-Page-green)](https://reward-forcing.github.io/)
[![Models](https://img.shields.io/badge/🤗-Models-yellow)](https://huggingface.co/JaydenLu666/Reward-Forcing-T2V-1.3B)

</div>

## 🚀 Progress

- [x] 📝 Technical Report / Paper
- [x] 🌐 Project Homepage
- [x] 💻 Training & Inference Code
- [x] 🤗 Pretrained Model: T2V-1.3B
- [ ] 🔜 Pretrained Model: T2V-14B (In progress)


## 🎯 Overview

<div align="center">
  <img src="assets/teaser.png" width="800px">
</div>

> **TL;DR**: We propose Reward Forcing to distill a bidirectional video diffusion model into a 4-step autoregressive student model that enables real-time (23.1 FPS) streaming video generation. Instead of using vanilla distribution matching distillation (DMD), Reward Forcing adopts a novel rewarded distribution matching distillation (Re-DMD) that prioritizes matching towards high-reward regions, leading to enhanced object motion dynamics and immersive scene navigation dynamics in generated videos.



## 📋 Table of Contents

- [Requirements](#-requirements)
- [Installation](#-installation)
- [Pretrained Checkpoints](#-pretrained-checkpoints)
- [Inference](#-inference)
- [Training](#-training)
- [Results](#-results)
- [Citation](#-citation)
- [Acknowledgements](#-acknowledgements)
- [Contact](#-contact)


## 🔧 Requirements

- GPU: NVIDIA GPU with at least 24GB memory for inference, 80GB memory for training.
- RAM: 64GB or more recommended.
- Linux operating system.

## 🛠️ Installation

### Step 1: Clone the repository
```bash
git clone https://github.com/JaydenLyh/Reward-Forcing.git
cd Reward-Forcing
```

### Step 2: Create conda environment
```bash
conda create -n reward_forcing python=3.10
conda activate reward_forcing
```


### Step 3: Install dependencies
```bash
pip install -r requirements.txt
pip install flash-attn --no-build-isolation
```

### Step 4: Install the package
```bash
pip install -e .
```

## 📦 Pretrained Checkpoints
### Download Links

| Model |  Download |
|-------|----------|
| VideoReward |  [Hugging Face](https://huggingface.co/KlingTeam/VideoReward) |
| Wan2.1-T2V-1.3B |  [Hugging Face](https://huggingface.co/Wan-AI/Wan2.1-T2V-1.3B) |
| Wan2.1-T2V-14B |  [Hugging Face](https://huggingface.co/Wan-AI/Wan2.1-T2V-14B) |
| ODE Initialization | [Hugging Face](https://huggingface.co/gdhe17/Self-Forcing/blob/main/checkpoints/ode_init.pt) |
| Reward Forcing | [Hugging Face](https://huggingface.co/JaydenLu666/Reward-Forcing-T2V-1.3B) |

### File Structure
After downloading, organize the checkpoints as follows:
```
checkpoints/
├── Videoreward/
│   ├── checkpoint-11352/
│   └── model_config.json
├── Wan2.1-T2V-1.3B/
├── Wan2.1-T2V-14B/
├── Reward-Forcing-T2V-1.3B/
└── ode_init.pt
```

### Quick Download Script
```bash
pip install "huggingface_hub[cli]"

# Download all checkpoints
bash download_checkpoints.sh
```


## 🚀 Inference
### Quick Start
```bash
# 5-seconds video inference
python inference.py \
    --num_output_frames 21 \
    --config_path configs/reward_forcing.yaml \
    --checkpoint_path checkpoints/Reward-Forcing-T2V-1.3B/rewardforcing.pt \
    --output_folder videos/rewardforcing-5s \
    --data_path prompts/MovieGenVideoBench_extended.txt \
    --use_ema

# 30-seconds video inference
python inference.py \
    --num_output_frames 120 \
    --config_path configs/reward_forcing.yaml \
    --checkpoint_path checkpoints/Reward-Forcing-T2V-1.3B/rewardforcing.pt \
    --output_folder videos/rewardforcing-30s \
    --data_path prompts/MovieGenVideoBench_extended.txt \
    --use_ema
```

## 🏋️ Training
### Multi-GPU Training
```bash
# bash train.sh
torchrun --nnodes=1 --nproc_per_node=8 --rdzv_id=5235 --rdzv_backend=c10d  \
    --rdzv_endpoint=$MASTER_PORT train.py  --config_path configs/reward_forcing.yaml \
    --logdir logs/reward_forcing \
    --disable-wandb
```

### Multi-Node Training
```bash
torchrun --nnodes=$NODE_SIZE --nproc_per_node=8 --node-rank=$NODE_RANK --rdzv_id=5235 --rdzv_backend=c10d  \
    --rdzv_endpoint=$MASTER_IP:$MASTER_PORT train.py  --config_path configs/reward_forcing.yaml \
    --logdir logs/reward_forcing \
    --disable-wandb
```

### Configuration Files
Training configurations are in `configs/`:
- `default_config.yaml`: Default configuration
- `reward_forcing.yaml`: Reward Forcing configuration



## 📊 Results
### Quantitative Results

#### Performance on VBench
| Method | Total Score | Quality Score | Semantic Score | Params | FPS |
|--------|----------|----------|----------|--------|-----|
| SkyReels-V2 | 82.67 | 84.70 | 74.53 | 1.3B | 0.49 |
| MAGI-1 | 79.18 | 82.04 | 67.74 | 4.5B | 0.19 |
| NOVA | 80.12 | 80.39 | 79.05 | 0.6B | 0.88 |
| Pyramid Flow | 81.72 | 84.74 | 69.62 | 2B | 6.7 |
| CausVid | 82.88 | 83.93 | 78.69 | 1.3B | 17.0 |
| Self Forcing | 83.80 | 84.59 | 80.64 | 1.3B | 17.0 |
| LongLive | 83.22 | 83.68 | **81.37** | 1.3B | 20.7 |
| **Ours** | **84.13** | **84.84** | 81.32 | 1.3B | **23.1** |


### Qualitative Results
Visualizations can be found in our [Project Page](https://reward-forcing.github.io/).




## 📄 Citation
If you find this work useful, please consider citing:

```bibtex
@article{lu2025reward,
  title={Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation},
  author={Lu, Yunhong and Zeng, Yanhong and Li, Haobo and Ouyang, Hao and Wang, Qiuyu and Cheng, Ka Leong and Zhu, Jiapeng and Cao, Hengyuan and Zhang, Zhipeng and Zhu, Xing and others},
  journal={arXiv preprint arXiv:2512.04678},
  year={2025}
}
```


## 🙏 Acknowledgements
This project is built upon several excellent works: [CausVid](https://github.com/tianweiy/CausVid), [Self Forcing](https://github.com/guandeh17/Self-Forcing), [Infinite Forcing](https://github.com/SOTAMak1r/Infinite-Forcing), [Wan2.1](https://github.com/Wan-Video/Wan2.1), [VideoAlign](https://github.com/KlingTeam/VideoAlign)

We thank the authors for their great work and open-source contribution.


## 📧 Contact
For questions and discussions, please:
- Open an issue on [GitHub Issues](https://github.com/JaydenLyh/Reward-Forcing/issues)
- Contact us at: yunhonglu@zju.edu.cn