Discover Agent Party: Your Ultimate 3D AI Desktop Companion – Complete Guide to Features, Installation, and Usage

Have you ever imagined having an AI desktop companion that can chat with you, control your smart home devices, and even deploy seamlessly to platforms like WeChat and QQ? Meet Agent Party – a powerful, versatile 3D AI desktop companion that redefines what’s possible with artificial intelligence. This innovative tool integrates enterprise-level capabilities like knowledge base integration, real-time internet access, permanent memory, and multi-modal interaction, all while supporting cross-platform deployment.

What is Agent Party?

Agent Party is an open-source 3D AI desktop companion project designed to provide users with a highly customizable, feature-rich intelligent agent experience. It supports multiple modular functions that enhance existing language models without requiring coding expertise, making it suitable for personal entertainment, enterprise services, development testing, and various other scenarios.

Key Feature Highlights

How to Install Agent Party

Agent Party offers multiple installation methods suitable for both regular users and developers.

🖥️ Windows Installation Guide

Windows users can choose between installed or portable versions:

• Installed Version: Download Installer
  During installation, select “Install for current user only” to avoid requiring administrator privileges when launching.

• Portable Version: Download Portable Package
  Extract and double-click start.bat to run, or use update.bat to update the software.

🍎 macOS Installation Guide

1. Download DMG File
2. Drag the App file to your Applications folder
3. Open Terminal and execute this command to remove security restrictions:

sudo xattr -dr com.apple.quarantine /Applications/Super-Agent-Party.app

🐧 Linux Installation Guide

Linux users have two installation options:

• AppImage Format: Download AppImage File
  Grant execution permissions and run directly.

• DEB Package (For Ubuntu/Debian): Download DEB Package
  Install using your package manager.

🐳 Docker Deployment (Recommended for Server Environments)

If you prefer Docker, use these commands for quick deployment:

docker pull ailm32442/super-agent-party:latest
docker run -d -p 3456:3456 -v ./super-agent-data:/app/data ailm32442/super-agent-party:latest

After deployment, access the application at http://localhost:3456/. All data is stored locally in the ./super-agent-data directory, ensuring privacy and security.

📦 Source Code Deployment (For Developers)

If you want to build from source, follow these steps:

Windows System:

git clone https://github.com/heshengtao/super-agent-party.git
cd super-agent-party
uv sync
npm install
start_with_dev.bat

Linux/Mac System:

git clone https://github.com/heshengtao/super-agent-party.git
cd super-agent-party
uv sync
npm install
chmod +x start_with_dev.sh
./start_with_dev.sh

Core Features of Agent Party

1. Multi-Provider Support

Agent Party allows connection to various model services, whether locally deployed (Ollama, Dify) or cloud-based APIs, all integrated seamlessly.

2. Rich Tool Ecosystem

Built-in tools include knowledge bases, web search, smart home control, browser operations, and more. All tools support asynchronous calls without blocking AI responses.

3. VRM Virtual Desktop Companion

Upload custom VRM models to create unique desktop companion avatars for more engaging interactions.

4. Character Cards and Memory System

Supports importing Tavern character cards with long-term memory capabilities for more personalized AI interactions.

5. WeChat/QQ Bot Deployment

Deploy your agent to WeChat or QQ with one click, enabling non-intrusive chatbot services while avoiding account restrictions.

6. Developer-Friendly Interfaces

Provides OpenAI API-compatible interfaces and MCP protocol support for easy secondary development and integration.

7. ComfyUI Workflow Integration

Convert ComfyUI workflows into agent tools with support for multi-server load balancing.

System Requirements and Compatibility

Agent Party is designed to be lightweight with minimal hardware requirements:

• CPU: 2 cores or more
• RAM: 2GB or more

Since all models are optional, you can use cloud services exclusively or run lightweight models locally. It runs smoothly even on 2-core, 2GB cloud servers via Docker deployment.

How to Use Agent Party?

Basic Usage

• Desktop Version: Click the icon to launch after installation
• Web/Docker Version: Access via http://localhost:3456/ after startup
• API Calls: Developers can integrate using the following code:

from openai import OpenAI

client = OpenAI(
  api_key="super-secret-key",
  base_url="http://localhost:3456/v1"
)

response = client.chat.completions.create(
  model="super-model",
  messages=[
      {"role": "user", "content": "What is Agent Party?"}
  ]
)

print(response.choices[0].message.content)

MCP Protocol Integration

You can integrate Agent Party with other MCP-supported systems by adding this configuration:

{
  "mcpServers": {
    "super-agent-party": {
      "url": "http://127.0.0.1:3456/mcp"
    }
  }
}

Frequently Asked Questions (FAQ)

❓ Is Agent Party free to use?

Yes, Agent Party is an open-source project under AGPLv3 license, free for personal use. Commercial use in closed-source projects requires a commercial license.

❓ Does it support custom models?

Yes, you can connect locally deployed models (like Ollama) or cloud APIs (like OpenAI, Dify, etc.).

❓ Is my data uploaded to servers?

No, all data is stored locally unless you explicitly configure cloud services.

❓ Does it support voice interaction?

Yes, Agent Party supports multi-modal capabilities including voice input and output.

❓ How to update to newer versions?

Windows portable versions can update via update.bat. Other versions should check the project Release page for updates.

Conclusion and Support

Agent Party stands out as a comprehensive AI desktop companion that combines multiple powerful features for learning, development, and entertainment. Its modular design and cross-platform support make it one of the most flexible AI application frameworks available today.