IAMFlow-FP8

<p align="center" style="border-radius: 10px"> <img src="assets/icon+name.png" width="50%" alt="logo"/> </p>

<div align="center" >Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory<div align="center">

<div align="center"> <p> <a href="https://eddie0521.github.io/">Jinzhuo Liu</a>¹, <a href="https://zhangzjn.github.io">Jiangning Zhang</a>^{1<a href="mailto:186368@zju.edu.cn">✉</a>}, <a href="https://github.com/Rinke02">Wencan Jiang</a>¹, <a href="https://scholar.google.com/citations?user=xiK4nFUAAAAJ&hl=zh-CN">Yabiao Wang</a>², <a href="https://dk-liang.github.io/">Dingkang Liang</a>³, <a href="https://scholar.google.com/citations?user=m3KDreEAAAAJ&hl=en">Zhucun Xue</a>¹, <a href="https://yiranran.github.io/">Ran Yi</a>⁴, <a href="https://person.zju.edu.cn/yongliu">Yong Liu</a>¹ </p> <p> ¹Zhejiang University, &nbsp;&nbsp; ²Tencent Youtu Lab, &nbsp;&nbsp; ³Huazhong University of Science and Technology,<br> ⁴Shanghai Jiao Tong University &nbsp;&nbsp; ^{<a href="mailto:186368@zju.edu.cn">✉</a>}Corresponding author </p> </div> <p align="center"> <a href="https://eddie0521.github.io/projects/iamflow/"><img src="https://img.shields.io/badge/Project-Page-Green"></a> &nbsp; <a href="https://arxiv.org/abs/2605.18733"><img src="https://img.shields.io/static/v1?label=arXiv&message=2605.18733&color=red&logo=arxiv"></a> &nbsp; <a href="https://huggingface.co/Eddie0521/IAMFlow-FP8"><img src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Model-orange"></a> </p>

🔥 Updates

• [2026.05.15]: We release the github repo, the project page, the quantized model checkpoints, the NarraStream-Bench, and the paper.

📷 Introduction

💡TL;DR: IAMFlow uses explicit identity-aware memory to keep identities consistent across evolving narrative prompts, achieving faster and stronger long video generation on NarraStream-Bench.

✨ Highlights

1. We introduce IAMFlow, a training-free identity-aware memory framework that explicitly organizes historical information around persistent entities and attributes, enabling reliable identity preservation across evolving prompt transitions.
2. We design a systematic inference acceleration pipeline to make the framework computationally practical, combining asynchronous visual verification, adaptive prompt transition, and model quantization to preserve long-term consistency without sacrificing generation speed.
3. We introduce NarraStream-Bench, a modern benchmark suite for assessing long-term consistency in narrative streaming video generation. Extensive experiments and ablation studies demonstrate that IAMFlow achieves superior performance across various metrics while enabling more efficient inference.

🛠️ Installation

1. Install Requirements

git clone git@github.com:Eddie0521/IAMFlow.git
cd IAMFlow
conda create -n iamflow python=3.12 -y
conda activate iamflow

# Install PyTorch first according to your CUDA environment.
python -m pip install torch==2.9.1 torchvision==0.24.1
python -m pip install -r requirements.txt
pip install flash-attn --no-build-isolation

2. Download Checkpoints

Download models using hf:

pip install "huggingface_hub[cli]"
hf download Wan-AI/Wan2.1-T2V-1.3B --local-dir pretrained/Wan2.1-T2V-1.3B
hf download Eddie0521/IAMFlow --local-dir pretrained/iamflow_models
hf download Qwen/Qwen3-VL-2B-Instruct --local-dir pretrained/Qwen3-VL-2B-Instruct
hf download Qwen/Qwen3-4B-Instruct-2507 --local-dir pretrained/Qwen3-4B-Instruct-2507

🔑 Inference

We deploy DiT, TextEncoder, and LLM on one GPU, while VAE and VLM are deployed on another GPU.

bash ./scripts/run_iamflow.sh

📏 Evaluation & Benchmark

See the NarraStream-Bench.

🤗 Acknowledgement

• MemFlow: the codebase we built upon. Thanks for their wonderful work.
• Self-Forcing: the algorithm we built upon. Thanks for their wonderful work.
• Wan: the base model we built upon. Thanks for their wonderful work.

🌟 Citation

Please leave us a star 🌟 and cite our paper if you find our work helpful.

@misc{liu2026advancingnarrativelongvideo,
      title={Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory}, 
      author={Jinzhuo Liu and Jiangning Zhang and Wencan Jiang and Yabiao Wang and Dingkang Liang and Zhucun Xue and Ran Yi and Yong Liu},
      year={2026},
      eprint={2605.18733},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2605.18733}, 
}