Skip to main content

🏠 Local Setup with Ollama

Set up WP LLM to run locally on your machine using Ollama for private, offline WordPress development with AI assistance.

Prerequisites

Before you begin, ensure you have the following:

Hardware Requirements

  • 8GB RAM minimum (16GB recommended)
  • 4GB free disk space for the model
  • Multi-core processor (4+ cores recommended)

Software Requirements

  • macOS 10.15+, Linux (Ubuntu 18.04+), or Windows 10+ with WSL2
  • Terminal/Command Prompt access
  • WordPress development environment (local or remote)

WordPress Knowledge

  • Basic familiarity with WordPress development
  • Understanding of PHP, HTML, CSS, and JavaScript
  • Experience with WordPress hooks, filters, and the plugin/theme system

Step 1: Install Ollama

Ollama is the local AI runtime that will run WP LLM on your machine.

macOS Installation

bash
curl -fsSL https://ollama.ai/install.sh | sh

Linux Installation

bash
curl -fsSL https://ollama.ai/install.sh | sh

Windows Installation

bash
# Install WSL2 first, then run:
curl -fsSL https://ollama.ai/install.sh | sh

Verify Installation

bash
ollama --version

You should see output like: ollama version 0.1.0

Step 2: Download the WP LLM Model

Download the specialized WordPress model:

bash
ollama pull wp-llm

This will download the ~4GB model file. The first download may take 10-20 minutes depending on your internet connection.

Progress indicators:

  • pulling manifest - Downloading model metadata
  • pulling layers - Downloading model layers
  • verifying sha256 - Verifying model integrity
  • writing manifest - Finalizing installation

Step 3: Run WP LLM Locally

Start an interactive session:

bash
ollama run wp-llm

You'll see a prompt like:

>>>

Now you can have a conversation with WP LLM:

>>> Create a custom post type for 'Portfolio Projects'

Single Request Mode

For one-off requests:

bash
ollama run wp-llm "Create a custom post type for 'Portfolio Projects'"

API Mode (For Integration)

Start the API server:

bash
ollama serve

Then make HTTP requests:

bash
curl -X POST http://localhost:11434/api/generate \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wp-llm",
    "prompt": "Create a custom post type for Portfolio Projects"
  }'

Step 4: Your First WordPress Code Generation

Let's test WP LLM with some real WordPress development tasks:

Example 1: Custom Post Type

bash
ollama run wp-llm "Create a custom post type for 'Portfolio Projects' with client name, project date, and technologies used"

Example 2: REST API Endpoint

bash
ollama run wp-llm "Create a REST API endpoint that returns all published portfolio projects with proper authentication"

Troubleshooting Common Issues

Model Not Found

Error: Error: model 'wp-llm' not found

Solution:

bash
# Verify the model is downloaded
ollama list

# If not listed, download again
ollama pull wp-llm

Out of Memory

Error: Error: out of memory

Solution:

  • Close other applications to free up RAM
  • Restart Ollama: ollama stop && ollama start
  • Consider upgrading to 16GB+ RAM

Slow Performance

Symptoms: Slow response times

Solutions:

  • Ensure you have 8GB+ free RAM
  • Close unnecessary applications
  • Use SSD storage for better I/O performance
  • Consider using a smaller model variant

Connection Refused

Error: Error: connection refused

Solution:

bash
# Restart Ollama service
ollama stop
ollama start

Permission Issues (Linux/macOS)

Error: Error: permission denied

Solution:

bash
# Add your user to the ollama group
sudo usermod -a -G ollama $USER

# Log out and back in, or restart your system

Next Steps

Immediate Next Steps

  1. ✨ Key Capabilities - Learn about the seven core capabilities
  2. Practice with examples - Try different WordPress development tasks
  3. Explore the API - Integrate WP LLM into your development workflow

Advanced Usage

  • Batch processing - Generate multiple files at once
  • Custom prompts - Create reusable prompt templates
  • Integration - Connect with your IDE or build tools

Best Practices

  • Be specific in your prompts
  • Provide context about your WordPress setup
  • Test generated code in a development environment
  • Review security aspects of generated code

Integration Examples

VS Code Integration

Install the Ollama extension and configure WP LLM as your default model.

Shell Integration

Add to your .bashrc or .zshrc:

bash
alias wp-llm='ollama run wp-llm'

Script Integration

Create automation scripts:

bash
#!/bin/bash
ollama run wp-llm "Create a custom post type for $1" > "$1-cpt.php"

🎉 You're ready to start generating WordPress code with AI!

Try asking WP LLM to create something for your next WordPress project, or explore the Key Capabilities to see what's possible.