広告掲載
MCP.so
ログイン

MCP Multi Model

@K1vin1906

リポジトリ

基本情報

カテゴリ

AI とエージェント

ライセンス

MIT

トランスポート

stdio

作者

K1vin1906

投稿者

K1vin Gao

MCP Multi Model について

MCP server for Claude Code — parallel multi-model queries, smart routing, image/video gen across OpenAI (GPT-5, GPT Image), Gemini (Imagen 4, Veo), DeepSeek, Kimi & 12+ providers

MCP Multi Model の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "multi-model": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-multi-model"
      ],
      "env": {
        "DEEPSEEK_API_KEY": "<YOUR_KEY>",
        "GEMINI_API_KEY": "<YOUR_KEY>",
        "KIMI_API_KEY": "<YOUR_KEY>"
      }
    }
  }
}

MCP Multi Model のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Mcp Multi Model?

Mcp Multi Model is a single MCP server that gives Claude Code access to multiple LLM providers (OpenAI, Gemini, DeepSeek, Kimi, Grok, Perplexity, local models, and more) for image/video generation, web search, smart model routing, and side-by-side comparisons. It is designed for developers who want to use the best model for each task without switching tabs or managing separate integrations.

How to use Mcp Multi Model?

Install via npx mcp-multi-model or clone the repository and run npm install && npm run setup. Add the server to your Claude Code MCP config (e.g., ~/.mcp.json) with the required API keys in the env block. Optionally copy config.example.yaml to config.yaml to customize models, endpoints, and defaults.

Key features of Mcp Multi Model

  • Smart routing – auto-picks the best model for the task
  • Image generation via Gemini Imagen and OpenAI GPT Image
  • Video generation via Gemini Veo 3.1
  • Built-in web search (Google Search grounding, Kimi Chinese web)
  • Compare models side by side with ask_all or ask_both
  • One-line install with npx, no build step
  • Supports local models (Ollama, LM Studio, llama.cpp, vLLM)
  • Auto-retry, fallback, cost tracking, and response caching

Use cases of Mcp Multi Model

  • Generate images and videos directly from the terminal without leaving Claude Code
  • Research topics using web search grounded models (Gemini, Kimi)
  • Compare answers from multiple models to choose the best solution
  • Translate between Chinese and English using dedicated tools
  • Write code cheaply by routing to DeepSeek or local models

よくある質問

1

What models are supported?

12+ providers are preconfigured, including OpenAI, Gemini, DeepSeek, Kimi, Grok, Perplexity, Anthropic (via OpenRouter), Mistral, Groq, Qwen, GLM, Together, and local runners (Ollama, LM Studio, llama.cpp, vLLM). Models without an API key are skipped automatically.

2

How do I install Mcp Multi Model?

Add the server to your MCP config with `"command": "npx", "args": ["-y", "mcp-multi-model"]` and set your API keys in the `env` block. Alternatively, clone the repo and run `npm install && npm run setup` for an interactive wizard.

3

Do I need API keys?

Yes, for cloud models you need the corresponding API keys (e.g., `DEEPSEEK_API_KEY`, `GEMINI_API_KEY`). Local models (Ollama, LM Studio, etc.) require no API key and run entirely on your machine.

4

Is my data sent to the server author?

No. Mcp Multi Model is a local relay. No telemetry, analytics, or data is sent to the author. Prompts go directly from your machine to the LLM provider you configured.

5

How do I add a new model?

Add one block to your `config.yaml` file specifying the model name, adapter, endpoint, API key environment variable, and optional pricing/fallback settings. The server dynamically generates tools from your config.

コメント

「AI とエージェント」の他のコンテンツ

Rumoro

@Rumoro

Social listening for AI agents. Rumoro's remote MCP server lets Claude, Cursor, Codex and other MCP clients search and sort mentions of your product, your competitors and your topics on Reddit, X, Hacker News, GitHub, Yo

Cartesia MCP

cartesia-ai

Cartesia MCP is a remote MCP server for Cartesia (https://cartesia.ai) available at https://mcp.cartesia.ai.

scvd.store

seancrecord

scvd.store is an evidence observatory for agentic commerce. Independent signed observation of what other people's endpoints, artifacts and payments actually did among other things. USDC on Base, Arbitrum One, World Chain

2

TruVerifAI Panel-review

TruVerifAI

TruVerifAI's Panel Review is the guardian agent for AI's highest-stakes coding decisions: four frontier models argue over an AI coding agent's riskiest designs, diffs, and commits before they execute, with local review g

1

Hostinger

hostinger

Model Context Protocol (MCP) server for Hostinger API. If you don't want to install or run anything locally, connect directly to Hostinger's hosted MCP server: This opens a browser window to authorize via OAuth. Once app

148
広告掲載