---
title: Adversarial-Gradient-Forcing
canonical_url: "https://www.modelscope.cn/models/Wuli-Art/Adversarial-Gradient-Forcing"
md_url: "https://www.modelscope.cn/models/Wuli-Art/Adversarial-Gradient-Forcing.md"
repository: Wuli-Art/Adversarial-Gradient-Forcing
last_updated: 2026-10-14
license: apache-2.0
pipeline_tag: text-to-video-synthesis
tasks:
- text-to-video-synthesis
base_model:
- Wan-AI/Wan2.1-T2V-1.3B
- gdhe17/Self-Forcing
base_model_relation: finetune
library_name:
- pytorch
frameworks:
- pytorch
language:
- en
downloads: 0
stars: 0
tags:
- video-generation
- autoregressive
- diffusion
- adversarial-gradient-forcing
- "arxiv:2610.14553"
---
# Adversarial-Gradient-Forcing
> Adversarial-Gradient-Forcing - Wuli-Art 在 ModelScope 开源的模型。Adversarial Gradient Forcing Robust Few-Step Autoregressive Video Generation Zhejiang University · Alibaba Group · Nanjing University
Wuli-Art/Adversarial-Gradient-Forcing 是 ModelScope 魔搭社区上的text-to-video-synthesis模型，采用 apache-2.0 许可，基于 Wan-AI/Wan2.1-T2V-1.3B、gdhe17/Self-Forcing 构建。
- **Repository**: Wuli-Art/Adversarial-Gradient-Forcing
- **License**: apache-2.0
- **Tasks**: text-to-video-synthesis
- **Base model**: Wan-AI/Wan2.1-T2V-1.3B, gdhe17/Self-Forcing
- **Tags**: video-generation, autoregressive, diffusion, adversarial-gradient-forcing, arxiv:2610.14553
- **Downloads**: 0
- **Stars**: 0
- **Last updated**: 2026-10-14
Source: https://www.modelscope.cn/models/Wuli-Art/Adversarial-Gradient-Forcing
---
Adversarial Gradient Forcing
Robust Few-Step Autoregressive Video Generation
Zhejiang University · Alibaba Group · Nanjing University
**AGF** improves long-video generation by training models to withstand **adversarial perturbations in their generated history**. Trained on five-second rollouts, it supports generation up to four minutes with the same four-step sampler.
Generated with AGF · Selected video excerpts
Full videos and comparisons ↗
1.3B
Wan2.1 backbone |
4 steps
Denoising per block |
Up to 4 min
Video generation |
## 🧩 Model details
AGF selects adversarial history perturbations using distribution-matching gradients from later video blocks, then trains jointly on clean and perturbed histories. A first-step prediction consistency regularizer helps preserve scene structure during the initial training stage.
| Property | Paper checkpoint |
| --- | --- |
| Architecture | Wan2.1 1.3B autoregressive video generator |
| Initialization | Self-Forcing ODE initialization |
| Training | 200 regularized AGF + 200 pure AGF outer iterations |
| Selected weights | Global step 400, `generator_ema` |
| Training rollouts | 5 seconds; effective batch size 64 |
| Training prompts | Filtered, LLM-extended VidProM prompts used by Self-Forcing |
| Inference | 832 × 480, 16 fps, four denoising steps per block |
Iterations count outer training loops; the full schedule contains **80 generator updates**.
## 🚀 Download and inference
The paper checkpoint is available as [`model.pt`](model.pt) (17.03 GB). Download it with the Hugging Face CLI:
```bash
hf download Wuli-art/Adversarial-Gradient-Forcing model.pt \
--local-dir checkpoints/AGF
```
Follow the [installation and inference guide](https://github.com/wuli-art/Adversarial-Gradient-Forcing/blob/main/docs/inference.md) for the Python/CUDA environment and Wan 1.3B assets. Place the downloaded checkpoint at `checkpoints/AGF/model.pt` and run from the code repository root:
```bash
AGF_RUN_ID=preview bash scripts/infer.sh \
--checkpoint checkpoints/AGF/model.pt \
--prompts prompts/examples.txt --latent-frames 21 --gpus 0
```
The script selects **`generator_ema` by default**, matching the paper. Use `--latent-frames 21`, `240`, or `960` for approximately **5 seconds**, **1 minute**, or **4 minutes**. The default policy uses 12 latent frames of local attention, including 3 persistent sink frames, with relative sink positions enabled. See the [training guide](https://github.com/wuli-art/Adversarial-Gradient-Forcing/blob/main/docs/training.md) for the full training and checkpoint-resumption workflow.
## 📊 Results
Paper results under the shared evaluation protocol (higher is better). The 5-second and 60-second evaluations use 944 and 240 VBench prompts, respectively; the 240-second evaluation uses 120 MovieGenVideoBench extended prompts.
| Model | 5 s · VBench Total | 60 s · VBench Total | 240 s · Quality 5 |
| --- | ---: | ---: | ---: |
| **AGF** | **82.03** | **81.47** | **84.82** |
Quality 5 (Q5) is the arithmetic mean of subject consistency, background consistency, motion smoothness, aesthetic quality, and imaging quality. It excludes dynamic degree and is not an official VBench Total score. See the [evaluation protocol](https://github.com/wuli-art/Adversarial-Gradient-Forcing/blob/main/docs/evaluation.md) for metric definitions and reproducibility details.
## Intended use and limitations
AGF is intended for text-to-video generation and research on autoregressive video models. Long sequences may exhibit visual artifacts, motion inconsistencies, or prompt drift.
## 📝 Citation
```bibtex
@misc{xia2026adversarial,
title = {Adversarial Gradient Forcing for Robust Few-Step Autoregressive Video Generation},
author = {Chenxi Xia and Yixuan Xu and Shan Liu and Jiapeng Wang
and Maohua Li and Hanlin Tang and Kan Liu and Tao Lan
and Weiyi Lu and Xinggao Liu},
year = {2026},
eprint = {2610.14553},
archivePrefix = {arXiv},
primaryClass = {cs.CV},
url = {https://arxiv.org/abs/2610.14553}
}
```
License: Apache 2.0. Third-party models, datasets, and other assets retain their respective licenses.
