MCP.so
ログイン

Ollama MCP Chat

@godstale

Ollama MCP Chat について

Ollama MCP Chat is a desktop chatbot application that integrates Ollama's local LLM models with MCP (Model Context Protocol) servers

基本情報

カテゴリ

AI とエージェント

ライセンス

MIT

ランタイム

python

トランスポート

stdio

公開者

godstale

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": {
    "ollama-mcp-chat": {
      "command": "uv",
      "args": [
        "sync"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Ollama MCP Chat?

Ollama MCP Chat is a desktop chatbot application that integrates Ollama’s local LLM models with MCP (Model Context Protocol) servers. It provides a Python / PySide6-based GUI, supports various tool calls, and allows users to freely extend capabilities via MCP servers.

How to use Ollama MCP Chat?

Clone the repository, install uv, run uv sync, install Ollama and pull a model (e.g., ollama pull qwen3:14b). Optionally configure MCP servers in mcp_config.json. Launch the GUI with uv run main.py. Type quit, exit, or bye to exit.

Key features of Ollama MCP Chat

  • Run Ollama LLM models locally for free
  • Integrate and call various tools via MCP servers
  • Manage and save chat history
  • Real-time streaming responses and tool call results
  • Intuitive desktop GUI (PySide6-based)
  • GUI support for adding, editing, and removing MCP servers

Use cases of Ollama MCP Chat

  • Local AI assistant that can invoke external tools (e.g., weather, file operations)
  • Prototype and test MCP server integrations with an Ollama backend
  • Experiment with different LLM models and tool configurations
  • Learn how to build an AI desktop application with a Python GUI

FAQ from Ollama MCP Chat

What are the system requirements?

Python 3.12 or higher, Ollama installed (for local LLM execution), and uv (recommended for package management). MCP servers can be implemented in any language.

How do I add or modify MCP servers?

Add server information to the mcp_config.json file (e.g., command, arguments, transport). The GUI also supports adding, editing, and removing MCP servers directly.

Can I chat without configuring any MCP server?

Yes, basic LLM chat works even without an MCP server configuration.

Where is my chat history stored?

All conversations are automatically saved to chat_history.json. You can load previous chats or start a new chat from the GUI.

How do I exit the application?

Type quit, exit, or bye in the program to exit.

コメント

「AI とエージェント」の他のコンテンツ