Live2D Assistant

Name: Live2D Assistant
Author: wwbweibo

@wwbweibo

Repository

About Live2D Assistant

Live2d Assistant is an extensive assistant application with mcp server and multi-agent support.

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "live2d-assistant": {
      "command": "uv",
      "args": [
        "sync"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Live2D Assistant?

Live2D Assistant is a multi-platform (Web/Electron) assistant that integrates Live2D virtual avatars with AI intelligent dialogue. It supports custom Live2D models and multiple large language models (LLMs), making it suitable for interactive anime-style assistants, AI companions, and smart desktop setups.

How to use Live2D Assistant?

Install the frontend (cd live2d-assistant-fe, npm install, npm run dev), the backend (cd live2d-assistant-server, uv sync, uv run fastapi run main.py), and optionally the Electron desktop app (cd electron-live2d, npm install, npm run start). The frontend dev server runs on localhost:5173, backend on port 8000.

Key features of Live2D Assistant

Custom Live2D models with adjustable position, scale, and background.
Multi-LLM integration (Qwen, OpenAI, Gemini, etc.) with context-aware conversations.
Multi-agent system using Router Agent and expert agents for task delegation.
Web (Vite+Vue3) and Electron desktop support, switchable in one codebase.
Rich settings including assistant name, system prompt, model parameters, and background.

Use cases of Live2D Assistant

A customizable anime-style virtual assistant for daily desktop interaction.
An AI-powered Live2D character for live streaming or video calls.
A multi-agent chatbot that routes complex requests to specialized experts.
A personal AI companion with a persistent visual avatar on multiple platforms.

FAQ from Live2D Assistant

What LLMs are supported?

The assistant integrates multiple LLMs including Qwen, OpenAI, and Gemini, with more possible via configuration.

Can I use my own Live2D model?

Yes, the platform supports custom Live2D models along with custom backgrounds, scaling, and positioning.

What platforms does Live2D Assistant run on?

It supports Web (Vite+Vue3 development server) and Electron desktop applications, all from a single project.

How does the agent system work?

A Router Agent acts as the entry point, understands user intent, and distributes tasks to expert agents. Agents can transfer back to the Router for further delegation.

What are the runtime requirements?

Frontend and Electron require Node.js; the backend requires Python 3 and FastAPI with the uv project manager for dependencies.

Comments

More AI & Agents MCP servers

1Panel

1Panel-dev

🔥 1Panel is a modern, open-source VPS control panel — and the only one with native AI agent support. Run Ollama models, deploy OpenClaw agents, and manage your entire server stack from one clean web interface.

36.1K