Freepik Featured Image Generator

A WordPress plugin that generates AI-powered featured images for posts using Freepik or OpenRouter API. Supports multiple providers, AI models, customizable styles, and automatic generation.

Features

• Multiple API Providers: Freepik and OpenRouter with easy switching
• AI-Powered Prompts: OpenRouter uses a text LLM to analyze post content and generate optimal image prompts
• Multiple AI Models:
  • Freepik: Mystic, Flux Dev, Flux Pro 1.1, Flux 2 Pro, Flux 2 Turbo, HyperFlux
  • OpenRouter: Nano Banana 2 (Gemini 3.1 Flash), Nano Banana (Gemini 2.5 Flash), Nano Banana Pro (Gemini 3 Pro), GPT-5 Image
• Manual Style Selector: Choose style preset manually or auto-detect by category
• Customizable Styles: Define visual styles for different content categories
• Flexible Prompts: Edit prompt templates with placeholders (Freepik) or LLM system instructions (OpenRouter)
• Output Formats: Save images as WebP, AVIF, JPEG, or PNG
• Auto-Generation: Optionally generate images automatically when posts are published
• Multiple Post Types: Enable for posts, pages, or custom post types
• Metabox Interface: Easy-to-use interface in the post editor
• REST API: Programmatic access for custom integrations
• Anti-Text Protection: Built-in negative prompt to prevent text in images

Installation

1. Download the latest release
2. Go to WordPress Admin > Plugins > Add New > Upload Plugin
3. Upload the zip file and click "Install Now"
4. Activate the plugin
5. Go to Settings > AI Image Generator to configure

Configuration

API Provider

Choose between Freepik and OpenRouter in Settings > AI Image Generator > API Provider. The settings page dynamically shows configuration fields for the selected provider.

Freepik Setup

1. Get your API key from Freepik API
2. Enter the key in Freepik API Key field

OpenRouter Setup

1. Get your API key from OpenRouter
2. Enter the key in OpenRouter API Key field
3. Select an Image Model for generation
4. Select a Prompt Model (text LLM that analyzes your content)
5. Customize the Prompt System Instructions if needed

How OpenRouter Works

OpenRouter uses a two-step process:

1. Step 1 — Prompt Generation: A text LLM (e.g., Gemini 2.5 Flash) reads the full post content, title, category, and style preferences, then generates an optimized image generation prompt
2. Step 2 — Image Generation: The generated prompt is sent to an image model (e.g., Gemini 3.1 Flash) to create the featured image

This produces more contextually relevant images compared to template-based prompts.

AI Models

Freepik Models

Model	Best For	Quality	Speed
Mystic	Illustrations, flat design	Excellent	Medium
Flux Dev	General purpose	Good	Fast
Flux Pro 1.1	High quality	Premium	Slow
Flux 2 Pro	Latest premium	Premium	Medium
Flux 2 Turbo	Fast premium	Premium	Fast
HyperFlux	Speed priority	Good	Ultra Fast

OpenRouter Image Models

Model	Description
Nano Banana 2 (Gemini 3.1 Flash)	Latest, recommended
Nano Banana (Gemini 2.5 Flash)	Stable, fast
Nano Banana Pro (Gemini 3 Pro)	Higher quality
GPT-5 Image	OpenAI image generation

OpenRouter Prompt Models

Model	Best For
Gemini 2.5 Flash	Fast, cheap, recommended
Gemini 2.5 Pro	Higher quality prompts
Claude Sonnet 4	Creative prompts
GPT-4.1 Mini	Fast alternative

Output Formats

Format	Best For	File Size
WebP	Modern browsers (recommended)	Small
AVIF	Latest browsers	Smallest
JPEG	Maximum compatibility	Medium
PNG	Transparency support	Large

Aspect Ratios

• 1:1 Square - Social media posts
• 4:3 Classic - Traditional photos
• 3:2 Traditional - Standard photographs
• 2:1 Horizontal - Blog headers (recommended)
• 16:9 Widescreen - Video thumbnails
• 21:9 Panoramic - Wide banners
• 9:16 Vertical - Stories, mobile

System Prompt (Freepik)

Customize the prompt template using these placeholders:

Placeholder	Description
{title}	Post title
{style_description}	Style colors from category
{elements}	Visual elements from category
{mood}	Image mood from category
{category}	Category name
{excerpt}	Post excerpt (first 100 chars)

Category Styles

Define visual styles for each content category:

• Key: Category slug (e.g., "taxes", "marketing")
• Name: Display name
• Colors: Color palette description
• Elements: Visual elements to include
• Mood: Overall mood/feel

Usage

Post Editor

1. Create or edit a post
2. Find the "AI Image Generator" metabox in the sidebar
3. Select style manually or use "Auto (by category)"
4. Click "Generate Image" or "Regenerate"
5. Wait 10-30 seconds for generation

REST API

Generate image for a post:

POST /wp-json/fpfig/v1/generate/{post_id}
POST /wp-json/fpfig/v1/generate/{post_id}?style=marketing
POST /wp-json/fpfig/v1/generate/{post_id}?force=1

Generate with custom prompt:

POST /wp-json/fpfig/v1/custom
Content-Type: application/json

{
  "prompt": "Your custom prompt here",
  "post_id": 123  // Optional: attach to post
}

Requirements

• WordPress 5.8 or higher
• PHP 7.4 or higher
• GD library (for WebP/AVIF conversion)
• Freepik API key or OpenRouter API key

Changelog

1.3.0

• Added OpenRouter as second API provider
• AI-powered prompt generation: text LLM reads post content and creates optimized image prompts
• New image models: Gemini 3.1 Flash, Gemini 2.5 Flash, Gemini 3 Pro, GPT-5 Image
• New prompt models: Gemini 2.5 Flash, Gemini 2.5 Pro, Claude Sonnet 4, GPT-4.1 Mini
• Configurable prompt system instructions
• Dynamic settings page: shows/hides fields based on selected provider
• Provider-aware API test, metabox, and notices

1.2.3

• Added output quality slider for JPEG/WebP/AVIF
• Minor UI improvements

1.2.1

• Fixed Flux Pro API endpoint URL
• Added new models: Flux 2 Pro, Flux 2 Turbo, HyperFlux
• Removed non-existent Flux Realism model

1.2.0

• Added manual style selector in metabox
• Style preview updates dynamically
• Selected style saved per post

1.1.1

• Added negative_prompt to prevent text in images

1.1.0

• Added output format selection (WebP, AVIF, JPEG, PNG)
• Image conversion with GD library

1.0.0

• Initial release
• Support for Mystic, Flux Dev, Flux Pro models
• Customizable styles and prompts
• Auto-generation option
• REST API endpoints

License

GPL v2 or later

Credits

Developed by Sergey Nesmachny

Uses Freepik AI API and OpenRouter API for image generation.