Ollama Local Setup

Ollama lets you run AI models locally on your machine. This guide covers installation, model selection, and OmniIDE configuration.

Install Ollama

1. Download from ollama.ai
2. Run the installer
3. Verify installation: ollama --version

Pull a Model

# Recommended for coding
ollama pull codellama:13b

# Lighter alternative
ollama pull codellama:7b

# Start the server
ollama serve

Configure OmniIDE

{
  "omni.ollama.enabled": true,
  "omni.ollama.endpoint": "http://localhost:11434",
  "omni.ollama.model": "codellama:13b"
}

Related

• Ollama Integration
• Offline Mode