MCP (Model Context Protocol) 서버

@sangminpark9

About MCP (Model Context Protocol) 서버

No overview available yet

Config

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

{
  "mcpServers": {
    "mcp-server-sangminpark9": {
      "command": "python",
      "args": [
        "setup.py"
      ]
    }
  }
}

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 MCP (Model Context Protocol) 서버?

MCP (Model Context Protocol) 서버 is a service that integrates and manages multiple large language models (DeepSeek and Llama) through a standardized API interface. It provides easy model switching, conversation context management, and supports multiple storage backends. It is designed for developers who need a unified interface to work with different LLMs.

How to use MCP (Model Context Protocol) 서버?

Clone the repository, run python setup.py (optionally with --download-models) to set up the environment, and start the server with uvicorn app.main:app --reload. Docker users can run docker-compose up -d. The main API endpoint is POST /api/chat with a JSON body containing session_id, messages, model, and parameters.

Key features of MCP (Model Context Protocol) 서버

Integrates DeepSeek and Llama models
Easy model switching and routing
Conversation context management
Standardized REST API interface
Multiple storage backends: memory, Redis, SQLite

Use cases of MCP (Model Context Protocol) 서버

Building chatbots that switch between DeepSeek and Llama
Text generation with customizable parameters
Korean‑language translation and reasoning tasks
Development of LLM‑powered applications with persistent session context

FAQ from MCP (Model Context Protocol) 서버

What Python version is required?

The server requires Python 3.8 or higher.

Can I run the server without a GPU?

Yes, a CPU‑only environment is supported, but a CUDA‑compatible GPU is recommended for better performance.

Which models are supported?

DeepSeek and Llama models are supported. DeepSeek offers text‑generation, code‑generation, and reasoning; Llama is Korean‑language specialized with translation capabilities.

How do I add a new model?

Create a new model class in app/models that inherits from BaseModel, register it in app/models/router.py, and update the routing configuration in app/core/config.py.

How can I delete a conversation session?

Send a DELETE /api/sessions/{session_id} request. The server responds with a success status and message confirming deletion.

Frequently asked questions