Unity Ollama

🚀 Unity MCP and Ollama Integration Guide

This guide details how to integrate local Large Language Models (LLMs) via Ollama using the MCP protocol in Unity. It enables features such as creating scene objects, attaching scripts, and performing other operations. The document covers installation steps, configuration instructions, usage methods, and troubleshooting.

🚀 Quick Start

Prerequisites

• Python: Version 3.10 or higher
• Ollama: Download and install from ollama.ai
• Unity Editor: Ensure it is correctly installed

Explicit Setup Instructions

Create a Project Directory

mkdir C:\PythonMCP
cd C:\PythonMCP

Clone the Repository

git clone https://github.com/ZundamonnoVRChatkaisetu/unity-mcp-ollama.git

Copy the Python Folder

copy unity-mcp-ollama\Python .

Create a Virtual Environment

python -m venv venv

Activate the Virtual Environment

venv\Scripts\activate

Install Dependencies

cd Python
pip install -e .

Start the Server

python server.py

💻 Usage Examples

Sample Prompts for the Chat Interface

• "Create a red cube at position (0, 1, 0)"
• "Add a sphere to the scene and apply a blue material"
• "List all objects in the current scene"
• "Write a simple movement script and attach it to the cube"

🔧 Technical Details

Component Status Indicators

• Python Server Status:

  • Green: Connected
  • Yellow: Connection issue
  • Red: Not connected

• Unity Bridge Status:

  • Running: Unity is listening for connection requests
  • Stopped: Unity's socket server is not running

• Ollama Status:

  • Connected: Successfully connected to the Ollama server
  • Not connected: Unable to connect to Ollama

Performance Considerations

The performance of local LLMs depends on hardware configuration:

• For the deepseek-r1:14b model, at least 12GB of video memory is recommended.
• For the gemma3:12b model, at least 10GB of video memory is recommended.
• CPU-only mode is possible but significantly slower.

🛠️ Troubleshooting

Common Issues

1. Python Server Not Connected

   • Ensure the Python server is running (python server.py).
   • Check the Python console for error messages.
   • Confirm that the Unity bridge is started.

2. Unity MCP Menu Not Found

   • Ensure the Editor scripts are correctly imported into the project.
   • Check the Unity console for error messages.
   • Try restarting Unity.

3. Ollama Connection Issues

   • Ensure the Ollama service is running (ollama serve).
   • Verify that the model is correctly loaded.
   • Check firewall settings to ensure port 11434 is not blocked.

4. MCP Command Execution Failed

   • Check the Python console for detailed error messages.
   • Ensure the Unity bridge is running properly.
   • Confirm that the prompt instructions are clear.

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

Thanks to the following projects for their contributions:

• Based on Justin P. Barnett's unity-mcp
• Using Ollama for local LLM integration