Ollama Pterodactyl Egg

This Pterodactyl egg allows you to run Ollama easily on your panel and deploy local AI models with full API access.

Features

• Run Ollama inside the Pterodactyl environment.

• Support for multiple models such as:

  • Qwen

  • LLaMA

  • Gemma

• Built-in API server (/api/generate).

• Automatic model startup.

• Configurable model via startup variable.

• Optimized installation script with fallback download sources.

• Lightweight and stable setup.

How It Works

The egg installs Ollama automatically, starts the API server, and runs the selected model defined in the startup variable.

Startup Configuration

• MODEL_NAME → Choose the model to run.

  • Default value: qwen2.5:3b

Use Cases

• AI chatbots

• Discord bots

• Web AI integrations

• Local OpenAI alternative

• Testing LLM models

Important Notes

• The first startup may take some time while the selected model is downloaded.

• Sufficient RAM is required depending on the model being used.

• CPU performance directly affects response speed.