Deepchat

DeepChat is a powerful cross - platform AI assistant that supports the deployment of multiple cloud LLMs and local models, and has rich chatbot functions and search extension capabilities.

Artificial intelligence chatbots Developer tools #AI Assistant #Multi - Model #Search Extension #MCP Support Local .TypeScript

rating : 0 points

downloads : 2.1K

update time : 2025-04-23

What is the MCP server?

The MCP server is a model controller platform that integrates multiple functions for managing and controlling local and cloud-based LLMs (Large Language Models). It allows users to easily configure and run models through a graphical interface without complex command-line operations.

How to use the MCP server?

Using the MCP server is very simple. After installation, you can add, manage, and run models through the graphical interface. It supports multiple protocols and services, including StreamableHTTP, SSE, Stdio, etc.

Applicable scenarios

The MCP server is suitable for scenarios that require efficient management and operation of multiple LLMs, such as enterprise internal knowledge base queries, multi-model integrated development, and complex business process automation.

Main Features

Local Model ManagementSupports downloading, deploying, and running local models, such as Ollama, through a graphical interface.

MCP Service SupportComes with a built-in Node.js runtime environment and supports npx-like services without additional configuration.

Tool Invocation SupportProvides a detailed tool invocation debugging window that automatically formats tool parameters and returned data.

Streaming Protocol SupportSupports multiple streaming protocols, such as StreamableHTTP, SSE, Stdio, etc.

Advantages and Limitations

Advantages

User-friendly configuration interface simplifies model management.

Supports multi-protocol and multi-model integration, improving development efficiency.

Built-in tool support reduces the need for secondary installations.

Rich visualization functions facilitate monitoring and debugging.

Limitations

Some advanced features may require a certain technical background.

Relies on the network environment, and the offline mode has limited functionality.

High resource consumption, suitable for high-performance devices.

How to Use

Install the MCP server

Download and install DeepChat, and ensure that the Node.js environment is installed.

Configure the model

Add model sources, such as Ollama, DeepSeek, etc., through the graphical interface.

Start the service

Run the MCP service to start managing and invoking models.

Usage Examples

Local Model ManagementManage local Ollama models through the MCP server.

Remote API Model ServiceConnect to a remote API model through the MCP server.

Frequently Asked Questions

How to add a new model source?

Does the MCP server support offline mode?

How to view the list of added models?

Related Resources

Official Documentation of the MCP Server

Detailed usage guide for the MCP server.

GitHub Code Repository

Open-source code repository for the MCP server.

Video Tutorial

Video tutorial for quickly getting started with the MCP server.

🚀 Deep Chat Project Introduction

This project offers a comprehensive chat solution with various features and broad model compatibility.

🚀 Quick Start

This deep chat project provides a range of powerful features and is highly compatible with multiple API formats. You can quickly start using it by following the installation and startup steps below.

✨ Features

Local Model Management: Supports local model deployment using Ollama.
Local File Handling: Capable of handling local files.
Artifacts Support: Provides functions related to Artifacts.
Custom Search Engine: Implemented through model parsing without the need for API adaptation.
MCP Support: Built - in npx functionality, no additional Node environment installation required.
Multimodal Model Compatibility: Supports multimodal models.
Chat Data Backup and Restoration: Offers backup and restoration functions for local chat data.
Compatibility with Model Providers: Compatible with any model provider following the OpenAI, Gemini, or Anthropic API formats.

📦 Installation

Install Dependencies

$ npm install
$ npm run installRuntime
# If an error occurs: No module named 'distutils'
$ pip install setuptools
# For Windows x64 systems
$ npm install --cpu=x64 --os=win32 sharp
# For Mac with Apple Silicon architecture
$ npm install --cpu=arm64 --os=darwin sharp
# For Mac with Intel architecture
$ npm install --cpu=x64 --os=darwin sharp
# For Linux x64 systems
$ npm install --cpu=x64 --os=linux sharp

Start Development

$ npm run dev

Build Guide

# Build for Windows systems
$ npm run build:win

# Build for macOS systems
$ npm run build:mac

# Build for Linux systems
$ npm run build:linux

# Build for specific architectures
$ npm run build:win:x64
$ npm run build:win:arm64
$ npm run build:mac:x64
$ npm run build:mac:arm64
$ npm run build:linux:x64
$ npm run build:linux:arm64

🤝 Contribution

If you're interested in contributing, please refer to the Contribution Guide. Windows and Linux packages are built by GitHub Actions.

📚 Documentation

The project has the following additional features:

Support for local model management (via Ollama).
Support for local file handling.
Provision of Artifacts - related functions.
Customizable search engine (parsed directly through the model without API adaptation).
Built - in MCP support (no additional Node environment required).
Compatibility with multimodal models.
Chat data backup and restoration functions.
Full compatibility with any model provider following the OpenAI, Gemini, and Anthropic API formats.