What is Qmcp

Qmcp is a powerful, cross-platform AI chat client built with Flutter, implementing the Model Context Protocol (MCP) for intelligent, context-aware interactions with multiple Large Language Models (LLMs).

Use cases

Use cases for Qmcp include automated customer service chatbots, virtual teaching assistants, content generation tools, and interactive storytelling applications.

How to use

To use Qmcp, install the necessary packages on your device, set up the environment according to the platform (Android, iOS, Desktop), and connect to your desired LLM. Follow the setup instructions provided in the README for detailed steps.

Key features

Key features of Qmcp include universal compatibility across major platforms, model flexibility allowing connection to any supported LLM, context awareness leveraging MCP for maintaining conversation context, and being enterprise-ready with a focus on security and scalability.

Where to use

Qmcp can be used in various fields such as customer support, education, content creation, and any domain requiring intelligent conversational agents that can interact with users in a context-aware manner.

Qubase MCP

A powerful, cross-platform AI Chat Client implementing the Model Context Protocol (MCP) for seamless AI interactions.

Overview

Qubase MCP is a sophisticated chat interface that revolutionizes AI interactions on desktop and mobile devices. Built with Flutter and implementing the Model Context Protocol, it provides a unified interface for multiple Large Language Models (LLMs) while ensuring secure, efficient, and context-aware conversations.

Features

Core Capabilities

Platform Support

Supported AI Models

• OpenAI (GPT-3.5, GPT-4)
• Anthropic Claude
• OLLama (Local Models)
• DeepSeek
• Custom Model Support

Local LLM Setup

Android Setup

1. Install Termux

   • Download Termux ARM64 V8 from Termux GitHub
   • Install and open Termux
   • Run termux-setup-storage to grant storage permissions
   • Run termux-change-repo to select package mirror
   • Update with pkg upgrade

2. Required Packages

   # Install Tur repository
   pkg install tur-repo
   
   # Install Ollama and Zellij
   pkg install ollama
   pkg install zellij
   

3. Android Configuration

   • Enable Developer Options: Settings > About device > Tap “Build number” 7 times
   • In Developer options, enable “Disable child process restrictions”

4. Model Operations

   # Start Ollama server
   ollama serve
   
   # In a new terminal, run models:
   ollama run deepseek-r1.5b  # For DeepSeek
   ollama run llama3.2        # For Llama3
   

5. Control Commands

   Action	Command
   Stop output	CTRL + C
   Exit model	CTRL + D
   Clear screen	CTRL + L
   Stop server	ps aux | grep ollama then kill [PID]

Desktop Setup

Coming soon…

System Requirements

Hardware Requirements

Software Requirements

Installation Guide

Prerequisites

1. Flutter Setup

   git clone https://github.com/flutter/flutter.git
   export PATH="$PATH:`pwd`/flutter/bin"
   flutter doctor
   

2. System Dependencies

   # Linux
   sudo apt-get update
   sudo apt-get install libsqlite3-0 libsqlite3-dev
   
   # macOS
   brew install sqlite3
   
   # Windows
   # SQLite included in Flutter Windows setup
   

3. Development Tools

   # Using uv (Recommended)
   brew install uv
   
   # Alternative: npm
   brew install node
   

Application Setup

1. Installation

   git clone https://github.com/qubasehq/Qmcp.git
   cd Qmcp
   flutter pub get
   

2. Launch

   # Desktop platforms
   flutter run -d <platform>  # macos, windows, linux
   
   # Mobile development
   flutter run -d <device-id>
   

Configuration

Initial Setup

1. API Configuration

   • Launch Qubase MCP
   • Navigate to Settings > API Configuration
   • Enter LLM API credentials
   • Configure custom endpoints (if needed)

2. MCP Server Setup

   • Access Settings > MCP Server
   • Choose installation method
   • Configure server settings
   • Select default AI model

File Locations

Reset Application

# Clear all data (use with caution)
rm -rf ~/Library/Application\ Support/run.daodao.qubase_mcp
rm -rf ~/Library/Application\ Support/qubase_mcp

Troubleshooting

For additional support, visit our Issues page.

Development Roadmap

Upcoming Features

Contributing

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to your branch
5. Open a Pull Request

Acknowledgments

• Model Context Protocol (MCP) Team
• MCP CLI Contributors
• Flutter Community
• Open Source AI Community

License

Licensed under Apache License 2.0 - see the LICENSE file for details.