> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Vidu - AI video generation

> Generate high-quality video from text, images, and reference materials using Vidu's Q1, Q2, and Q3 model generations in ComfyUI

Vidu is a fast AI video generation platform by Shengshu Technology, now available inside ComfyUI. Its three model generations — **Q1**, **Q2**, and **Q3** — deliver a range of speed and quality options, from foundation generation with broad input support to 3x faster inference with camera control and exceptional anime output driven by the latest Vidu engines (2.0, 3.0, and 3.5).

## Key capabilities

* **Multi-generation evolution** — Choose between Q1 (foundation), Q2 (fast + multi-reference + camera control), and Q3 (latest narrative + anime/2D focus) to match your workflow
* **Fast inference** — Q3 generates 4-second clips in as fast as 10 seconds; Q2 delivers 3x the speed of Q1
* **Rich input modes** — Text-to-video, image-to-video, reference-to-video (up to 7 images), start-end-to-video, video extension, and first-last-frame-to-video
* **Camera language (Q2)** — Pan, zoom, rotation, and other cinematic moves for precise shot direction
* **Exceptional anime & 2D animation (Q3)** — Industry-leading output quality for animation production pipelines
* **Subject consistency** — Maintain character and scene identity across multi-reference inputs

> [!TIP]
>
> 使用 API 节点需要保证你已经正常登录，并在受许可的网络环境下使用，请参考[API 节点总览](https://docs.comfy.org/zh/tutorials/partner-nodes/overview.md)部分文档来了解使用 API 节点的具体使用要求。

> [!TIP]
>
> > [tab] 本地用户
> >
> > 请确保你的 ComfyUI 已经更新。
> >
> > * [ComfyUI 下载](https://www.comfy.org/download)
> > * [ComfyUI 更新教程](https://docs.comfy.org/zh/installation/update_comfyui.md)
> >
> > 本指南里的工作流可以在[工作流模板](https://docs.comfy.org/zh/interface/features/template.md)中找到。如果找不到，可能是 ComfyUI 没有更新。
> >
> > 如果加载工作流时有节点缺失，可能原因有：
> >
> > 1. 你用的不是最新版（每夜版）。
> > 2. 启动时有些节点导入失败。
>
> > [tab] 云端用户
> >
> > * [Cloud](https://cloud.comfy.org) 会在 ComfyUI 稳定版本发布后更新。
> >
> > 所以，如果你发现本文档中有任何核心节点缺失，可能是因为新核心节点尚未在最新稳定版中发布。请等待下一个稳定版发布。

## Models

> [card] [Vidu Q1](https://docs.comfy.org/zh/tutorials/partner-nodes/vidu/vidu-q1.md)
>
> Foundation generation with text-to-video, image-to-video, reference-to-video and more — the versatile starting point.

> [card] [Vidu Q2](https://docs.comfy.org/zh/tutorials/partner-nodes/vidu/vidu-q2.md)
>
> 3x faster inference with multi-reference support and cinematic camera control — ideal for iterative workflows.

> [card] [Vidu Q3](https://docs.comfy.org/zh/tutorials/partner-nodes/vidu/vidu-q3.md)
>
> The latest generation with improved narrative understanding, outstanding anime/2D quality, and Vidu 3.5 engine.

> [card] [Vidu Q4 Preview](https://docs.comfy.org/zh/tutorials/partner-nodes/vidu/vidu-q4-preview.md)
>
> Newest preview generation alongside Q1–Q3.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.