How do I use Vidu MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Vidu MCP Generate a 5-second cinematic video of a golden retriever playing in autumn leaves." That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Vidu MCP

by shengshu-ai

Overview Schema Source Related Servers Discussions Admin

Python

Remote

Vidu MCP

A tool that allows you to access Vidu latest video generation models via applications that support the Model Context Protocol (MCP), such as Claude or Cursor.

中文文档

Overview

Vidu MCP is a tool that allows you to access Vidu latest video generation models via applications that support the Model Context Protocol (MCP), such as Claude or Cursor. This integration enables you to generate high-quality videos anytime, anywhere — including text-to-video, image-to-video, and more.

Related MCP server: @vidofy/mcp

Key Features

Text-to-Video Generation: Generate creative videos using text prompts
Image-to-Video Generation: Generate creative videos using text and image prompts
Reference-to-Video Generation: Generate creative videos using text and image prompts
StartEnd-to-Video Generation: Generate creative videos using text and image prompts

System Components

The system consists of two main components:

UVX MCP Server
- Python-based cloud server
- Communicates directly with the Vidu API
- Provides full video generation capabilities

Installation & Configuration

Prerequisites

Python 3.10 or higher
UV/UVX
Vidu API Key: Obtain from Vidu Platform (This feature requires API Credits, which must be purchased separately on Vidu Platform

Get Dependencies

Python:
- Download and install from the official Python website
- Ensure Python is added to your system path
UV/UVX:
- Install uv and set up our Python project and environment:

Mac/Linux

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

How to Use MCP Server

1. Get Vidu API Key

Visit the Vidu Platform
Register or log into your account
Create and copy your API key from the account settings

2. Download Required Dependencies

Python: Install Python 3.10 or above
UV/UVX: Install the latest stable version of UV & UVX

3. Configure MCP Client

Open your MCP client (e.g., Claude for Desktop or Cursor)
Locate the client settings
Open mcp_config.json (or relevant config file)
Add the configuration based on the method you use:

{
  "mcpServers": {
    "Vidu": {
      "command": "uvx",
      "args": [
        "vidu-mcp"
      ],
      "env": {
        "VIDU_API_KEY": "api-key-here", 
        "VIDU_API_HOST": "api-host-here"
      }
    }
  }
}

Save the config file

4. Restart MCP Client or Refresh MCP Server

Fully close and reopen your MCP client
Or use the "Refresh MCP Server" option if supported

Client-specific Configuration

Claude for Desktop

Open the Claude application
Navigate to Claude > Settings > Developer > Edit Config
Open the claude_desktop_config.json file
- Windows
- Mac : ~/Library/Application\ Support/Claude/claude_desktop_config.json
Add the configuration above and save
Restart Claude
- If connected successfully: the homepage will not show any error and the MCP status will be green
- If connection fails: an error message will be shown on the homepage

Cursor

Open the Cursor application
Go to Settings > Model Context Protocol
Add a new server
Fill in the server details as in the JSON config above
Save and restart or refresh the MCP server

Usage Examples

Text-to-Video

Use natural language prompts via Claude or Cursor to generate videos.

Basic Example:

In an ultra-realistic fashion photography style featuring light blue and pale amber tones, an astronaut in a spacesuit walks through the fog. The background consists of enchanting white and golden lights, creating a minimalist still life and an impressive panoramic scene.

Advanced Example with Parameters:

Generate a night cityscape video with the following parameters:
Prompt: Skyscraper lights twinkling under the night sky, with car lights forming streaks on the road
Model: viduq1
Style: general
Duration: 5 seconds
Aspect Ratio: 16:9
Resolution: 1080p
Movement Amplitude: middle

FAQ

How do I get a Vidu API key?

What should I do if the server doesn't respond?

Check whether your API key is valid
View error logs (typically in the log folders of Claude or Cursor)

How to obtain credits?

If you haven't topped up on the API platform yet, please do so first. Vidu Platform

Where can I find the generated video?

You will receive a URL link to view, download, or share the video.

How long does video generation take?

Typically 30 seconds to 5 minutes depending on complexity, server load, and network conditions.

What to do if you encounter a spawn uvx ENOENT error?

This error is typically caused by incorrect UV/UVX installation paths. You can resolve it as follows:

For Mac/Linux:

sudo cp ./uvx /usr/local/bin

For Windows:

Identify the installation path of UV/UVX by running the following command in the terminal:

where uvx

Open File Explorer and locate the uvx/uv files.
Move the files to one of the following directories:
- C:\Program Files (x86) or C:\Program Files

Support

Technical Support

Email: platform@vidu.studio
Website: https://platform.vidu.com

Available Tools

5 tools

generate_img_to_videoA

Generate a video from a pic and prompt.

COST WARNING: This tool makes an API call to Vidu which may incur costs. Only use when explicitly requested by the user.

Args:
    image (str, required): An image to be used as the start frame of the generated video
    model (str, required): The model to use. Values range ["viduq1","vidu1.5","vidu2.0"], with "viduq1" being the default.
    prompt (str, optional): A textual description for video generation, with a maximum length of 1500 characters
    duration (int, optional): Video duration. Default values vary by model:
                              - viduq1: default 5s, available: 5
                              - vidu2.0 and vidu1.5: default 4s, available: 4, 8
    seed (int, optional): Random seed
                          - Defaults to a random seed number
                          - Manually set values will override the default random seed
    resolution (str, optional): Resolution. Default values vary by model & duration:
                                - viduq1 (5s): default 1080p, available: 1080p
                                - vidu2.0 and vidu1.5 (4s): default 360p, options: 360p, 720p, 1080p
                                - vidu2.0 and vidu1.5 (8s): default 720p, options: 720p
    movement_amplitude (str, optional): The movement amplitude of objects in the frame.Defaults to auto, accepted value: auto small medium large
    bgm (bool, optional): Whether to add background music to the generated video.
                          - Default: false. Acceptable values: true, false.
                          - When true, the system will automatically add a suitable BGM.
                          - Only when the final generated video duration is 4 seconds is adding BGM supported.
Returns:
    task_id and video_url

Inputs: image (required), model, prompt, duration, +4 more.

generate_reference2video_to_videoA

Generate a video from a pic and prompt.

COST WARNING: This tool makes an API call to Vidu which may incur costs. Only use when explicitly requested by the user.

Args:
    images (str list, required): The model will use the provided images as references to generate a video with consistent subjects
                                 For fields that accept images:
                                 -Accepts 1 to 3 images.
    prompt (str, required): A textual description for video generation, with a maximum length of 1500 characters
    model (str, required): The model to use. Values range ["vidu1.5","vidu2.0"], with "vidu2.0" being the default.
    duration (int, optional): Video duration parameter, with default values depending on the model:
                              - vidu2.0: Default is 4 seconds, available option: 4
                              - vidu1.5: Default is 4 seconds, available options: 4, 8
    seed (int, optional): Random seed
                          - Defaults to a random seed number
                          - Manually set values will override the default random seed
    aspect_ratio (str, optional): The aspect ratio of the output video. Defaults to 16:9, accepted: 16:9 9:16 1:1
    resolution (str, optional): The resolution of the output video
                                Defaults to 360p , accepted value: 360p 720p 1080p
                                - Model vidu1.5 duration 4 accepted: 360p 720p 1080p
                                - Model vidu1.5 duration 8accepted: 720p
                                - Model vidu2.0 duration 4 accepted: 360p 720p
    movement_amplitude (str, optional): The movement amplitude of objects in the frame.Defaults to auto, accepted value: auto small medium large
    bgm (bool, optional): Whether to add background music to the generated video.
                          - Default: false. Acceptable values: true, false.
                          - When true, the system will automatically add a suitable BGM.
                          - Only when the final generated video duration is 4 seconds is adding BGM supported.
Returns:
    task_id and video_url

Inputs: images (required), prompt (required), model, duration, +5 more.

generate_startend2video_to_videoA

Generate a video from a pic and prompt.

COST WARNING: This tool makes an API call to Vidu which may incur costs. Only use when explicitly requested by the user.
Args:
    images (str list, required): Two images: first is start frame, second is end frame.
    model (str, required): The model to use. Values range ["vidu1.5","vidu2.0","viduq1-classic","viduq1"], with "viduq1" being the default.
    prompt (str, optional): A textual description for video generation, with a maximum length of 1500 characters
    duration (int, optional): Video duration. Default values vary by model:
                              - viduq1 and viduq1-classic: default 5s, available: 5
                              - vidu2.0 and vidu1.5: default 4s, available: 4, 8
    seed (int, optional): Random seed
                          - Defaults to a random seed number
                          - Manually set values will override the default random seed
    resolution (str, optional): Resolution (based on model & duration):
                                - viduq1 and viduq1-classic(5s): default 1080p, options: 1080p
                                - vidu2.0 and vidu1.5 (4s): default 360p, options: 360p, 720p, 1080p
                                - vidu2.0 and vidu1.5 (8s): default 720p, options: 720p
    movement_amplitude (str, optional): The movement amplitude of objects in the frame.Defaults to auto, accepted value: auto small medium large
    bgm (bool, optional): Whether to add background music to the generated video.
                          - Default: false. Acceptable values: true, false.
                          - When true, the system will automatically add a suitable BGM.
                          - Only when the final generated video duration is 4 seconds is adding BGM supported.
Returns:
    task_id and video_url

Inputs: images (required), model, prompt, duration, +4 more.

generate_template_to_videoA

Generate a video from a template.

COST WARNING: This tool makes an API call to Vidu which may incur costs. Only use when explicitly requested by the user.

Args:
    template (str, required): AI video template. Different templates have different call parameters.
    images (str list, required): Images
    prompt (str, optional): A textual description for video generation, with a maximum length of 1500 characters
    seed (int, optional): Random seed
                          - Defaults to a random seed number
                          - Manually set values will override the default random seed
    aspect_ratio (str, optional): The aspect ratio of the output video. Defaults to 16:9, accepted: 16:9 9:16 1:1
                                   - Different templates accepted different aspect ratio
    area (str, optional): Exotic Princess style control field only for template exotic_princess,
                          Default：auto, accepts：denmark,uk,africa,china,mexico,switzerland,russia,italy,korea,thailand,india,japan
    beast (str, optional): beast companion style control field only for template beast_companion，
                           Default auto, accepts：bear,tiger,elk,snake,lion,wolf
    bgm (bool, optional): Whether to add background music to the generated video.
                          - Default: false. Acceptable values: true, false.
                          - When true, the system will automatically add a suitable BGM.
                          - Only when the final generated video duration is 4 seconds is adding BGM supported.
Returns:
    task_id and video_url

Inputs: template (required), images (required), prompt, seed, +4 more.

generate_text_to_videoA

Generate a video from a prompt.

COST WARNING: This tool makes an API call to Vidu which may incur costs. Only use when explicitly requested by the user.

 Args:
    model (str, required): The model to use. Values range ["viduq1","vidu1.5"], with "viduq1" being the default.
    prompt (str, required): A textual description for video generation, with a maximum length of 1500 characters
    style (str, optional): The style of output video. Defaults to general, Accepted values: general anime
    duration (int, optional): Video duration. Default values vary by model:
                              - viduq1: default 5s, available: 5
                              - vidu1.5: default 4s, available: 4, 8
    seed (int, optional): Random seed
                          - Defaults to a random seed number
                          - Manually set values will override the default random seed
    aspect_ratio (str, optional): The aspect ratio of the output video. Values range ["1:1", "16:9","9:16"], with "16:9" being the default.
    resolution (str, optional): Resolution. Default values vary by model & duration:
                                - viduq1 (5s): default 1080p, available: 1080p
                                - vidu1.5 (4s): default 360p, available: 360p, 720p, 1080p
                                - vidu1.5 (8s): default 720p, available: 720p
    movement_amplitude (str, optional): The movement amplitude of objects in the frame.Defaults to auto, accepted value: auto small medium large
    bgm (bool, optional): Whether to add background music to the generated video.
                          - Default: false. Acceptable values: true, false.
                          - When true, the system will automatically add a suitable BGM.
                          - Only when the final generated video duration is 4 seconds is adding BGM supported.
Returns:
    task_id and video_url

Inputs: model, prompt, style, duration, +5 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.4/5.0

Scored 2026-04-18 06:55 across 5 tools

Disambiguation4/5

The tools are clearly differentiated by their input types and purposes: single image, multiple reference images, start-end images, template-based, and text-only generation. However, the first three tools have identical descriptions ('Generate a video from a pic and prompt'), which could cause initial confusion until the specific parameters are examined.

Naming Consistency5/5

All tools follow a consistent 'generate_[input_type]_to_video' naming pattern (e.g., generate_img_to_video, generate_text_to_video). The pattern is logical and predictable, making it easy to understand what each tool does based on its name alone.

Tool Count5/5

With 5 tools, this is well-scoped for a video generation server. Each tool covers a distinct input modality (image, reference images, start-end images, templates, text), providing comprehensive coverage without being overwhelming or sparse.

Completeness5/5

The tool set provides complete coverage of video generation modalities for this domain: from single images, multiple reference images, start-end sequences, templates, and pure text prompts. There are no obvious gaps in the core functionality, and all tools return consistent outputs (task_id and video_url).

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

MaxVideoAIA
com.maxvideoai
Plan, compare, price, generate, and recover AI video from compatible MCP clients.
AI & Machine Learning Image & Video Processing
321
https://api.maxvideoai.com/mcp
Writingmate MCP
ai.writingmate
Use AI models for chat, image, and video generation from Claude Code and other MCP hosts.
AI & Machine Learning Image & Video Processing
1
https://writingmate.ai/api/mcp
ViewMax MCPA
net.viewmaxstudio
Generate AI video from any MCP client. Pick the model, see the per-second price before you spend.
AI & Machine Learning Image & Video Processing
41
https://viewmaxstudio.net/api/mcp
TrezaA
io.github.treza-labs
Build, run, schedule, and publish AI video pipelines to YouTube and TikTok from any MCP client.
AI & Machine Learning Image & Video Processing Payments & Billing
391
https://www.trezalabs.com/api/mcp

Related MCP Servers

PixVerse MCPofficial
AI & Machine Learning Image & Video Processing Multimedia Processing
PixVerseAI
A
license
Not graded
quality
D
maintenance
Enables video generation from text, images, and more through MCP-compatible apps like Claude and Cursor.
Updated a year ago (2025-10-22 08:16 UTC)
55
MIT
@vidofy/mcpofficial
Image & Video Processing Audio Processing Speech Processing
vidofy
A
license
A
quality
B
maintenance
Enables generating images, video, audio, and speech from MCP clients using your own Vidofy account, with access to hundreds of models for text-to-video, image-to-video, image editing, lipsync, text-to-speech, and voice cloning.
Updated 16 days ago (2026-10-26 12:16 UTC)
9
101 npm
2
MIT
genrelay-mcp
Image & Video Processing AI & Machine Learning Multimedia Processing
gaoruitech
A
license
A
quality
B
maintenance
Enables video and image generation from any MCP client, with support for models like Veo 3.1, Grok Imagine, and Nano Banana, including text-to-video, image-to-video, image generation and editing.
Updated a month ago (2026-10-13 04:30 UTC)
4
81 npm
MIT
VIDIA MCP
Image & Video Processing Multimedia Processing AI & Machine Learning
lead788
A
license
Not graded
quality
C
maintenance
Enables AI agents to browse video production packages, obtain free quotes, start and track productions with budget safeguards, and download finished videos, thumbnails and subtitles from vidia.kr. It also supports account balance checks and asset library lookups through MCP clients such as Claude, Cursor and Codex.
Updated a month ago (2026-10-12 05:00 UTC)
65 npm
MIT

Vidu MCP

Overview

Key Features

System Components

Installation & Configuration

Prerequisites

Get Dependencies

Mac/Linux

Windows

How to Use MCP Server

1. Get Vidu API Key

2. Download Required Dependencies

3. Configure MCP Client

4. Restart MCP Client or Refresh MCP Server

Client-specific Configuration

Claude for Desktop

Cursor

Usage Examples

Text-to-Video

FAQ

Support

Technical Support

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

PixVerse MCPofficial

@vidofy/mcpofficial

genrelay-mcp

VIDIA MCP