---
title: Qwen-Video-Edit
canonical_url: "https://www.modelscope.cn/models/yunpeng1998/Qwen-Video-Edit"
md_url: "https://www.modelscope.cn/models/yunpeng1998/Qwen-Video-Edit.md"
repository: yunpeng1998/Qwen-Video-Edit
last_updated: 2026-08-21
parameters: 61.3B
tensor_type:
  - BF16
library_name:
  - safetensors
downloads: 163
stars: 0
---

# Qwen-Video-Edit

> Qwen-Video-Edit - yunpeng1998 在 ModelScope 开源的模型。This model is the instruction-based video editing checkpoint presented in Qwen-Video-Edit: Instruction-Based Video Editing by Repurposing an Image Editing Model. It repurposes an image editing model to…

- **Repository**: yunpeng1998/Qwen-Video-Edit
- **Parameters**: 61.3B
- **Downloads**: 163
- **Stars**: 0
- **Last updated**: 2026-08-21

Source: https://www.modelscope.cn/models/yunpeng1998/Qwen-Video-Edit

---

**Qwen-Video-Edit**

This model is the instruction-based video editing checkpoint presented in [Qwen-Video-Edit: Instruction-Based Video Editing by Repurposing an Image Editing Model](https://arxiv.org/abs/2608.14790). It repurposes an image editing model to directly edit video-VAE latents, requiring only lightweight trainable projections.

Project page: https://yunpeng1998.github.io/Qwen-Video-Edit-Page

Code: https://github.com/yunpeng1998/Qwen-Video-Edit


For usage instructions, please refer to the GitHub repository.