How Many Categories Are in the GPT-Image2 Prompt System: Complete 13-Category Breakdown

The GPT-Image2 prompt system organizes its style library into exactly 13 distinct categories, ranging from UI & Interfaces to Other Use Cases, as defined in the data/style-library.json file.

The freestylefly/awesome-gpt-image-2 repository provides a comprehensive style prompt library for GPT-Image2, OpenAI's advanced image generation model. Understanding the GPT-Image2 prompt system categories helps developers and designers quickly locate specific visual styles for their generative AI workflows. This analysis examines the source code structure to identify exactly how many categories exist and where they are defined.

Source File Location and Structure

The complete list of GPT-Image2 prompt system categories resides in data/style-library.json at the repository root. This JSON file contains an array of style objects, each containing a "category" field that assigns the prompt to one of 13 thematic groups.

When parsing the file, you can identify category boundaries by examining the line numbers where each unique category first appears in the source.

The 13 Categories of the GPT-Image2 Prompt System

Here is the complete enumeration of all categories as implemented in freestylefly/awesome-gpt-image-2:

1. UI & Interfaces – First defined at line 233
2. Charts & Infographics – First defined at line 260
3. Posters & Typography – First defined at line 314
4. Products & E-commerce – First defined at line 449
5. Brand & Logos – First defined at line 503
6. Architecture & Spaces – First defined at line 557
7. Photography & Realism – First defined at line 584
8. Illustration & Art – First defined at line 638
9. Characters & People – First defined at line 665
10. Scenes & Storytelling – First defined at line 719
11. History & Classical Themes – First defined at line 746
12. Documents & Publishing – First defined at line 773
13. Other Use Cases – First defined at line 800

Each category groups semantically related prompt templates that share structural similarities or visual domains.

How to Programmatically Count Categories

You can verify the category count programmatically using the following Node.js script. This approach reads the style-library.json file and aggregates unique values from the "category" fields:

import fs from 'fs';
import path from 'path';

const libPath = path.resolve('data', 'style-library.json');
const library = JSON.parse(fs.readFileSync(libPath, 'utf8'));

const categories = new Set();
library.forEach(item => categories.add(item.category));

console.log('Total categories:', categories.size);
console.log('Categories:', Array.from(categories));

Executing this script outputs:

Total categories: 13
Categories: [
  'UI & Interfaces',
  'Charts & Infographics',
  'Posters & Typography',
  'Products & E-commerce',
  'Brand & Logos',
  'Architecture & Spaces',
  'Photography & Realism',
  'Illustration & Art',
  'Characters & People',
  'Scenes & Storytelling',
  'History & Classical Themes',
  'Documents & Publishing',
  'Other Use Cases'
]

The Set data structure ensures uniqueness, preventing duplicate counting if multiple prompts share the same category string.

Supporting Files and Implementation

While data/style-library.json serves as the primary source of truth for GPT-Image2 prompt system categories, the repository includes additional files that reference these groupings:

• data/cases.json – Contains example use cases that map to the same 13 categories
• src/main.jsx – The front-end entry point that consumes the style library to render category filters in the UI

These files maintain consistency with the main style library, ensuring that any changes to the category taxonomy in style-library.json propagate correctly throughout the application.

Summary

• The GPT-Image2 prompt system contains exactly 13 distinct categories organized by visual domain and use case.
• Category definitions are stored in data/style-library.json at specific line numbers ranging from 233 to 800.
• Each style prompt object contains a "category" field that assigns it to one of the 13 groups.
• You can extract the complete category list programmatically using Node.js by aggregating unique values from the JSON array.
• Supporting files like cases.json and main.jsx rely on this categorical structure for UI rendering and example organization.

Frequently Asked Questions

How many categories are in the GPT-Image2 prompt system?

The system defines exactly 13 categories, including UI & Interfaces, Charts & Infographics, Photography & Realism, and Other Use Cases. This count is derived from the unique values in the "category" field of data/style-library.json.

Where is the category data stored in the repository?

All GPT-Image2 prompt system categories are stored in data/style-library.json in the repository root. Category assignments appear as string values in the "category" property of each style object, with the first occurrence of each category appearing between lines 233 and 800.

Can I add custom categories to the GPT-Image2 style library?

Yes, you can extend the existing taxonomy by adding new category strings to the "category" fields in data/style-library.json. However, to maintain compatibility with the front-end interface in src/main.jsx, you should also update any category filter logic that depends on the fixed enumeration of 13 groups.

What is the largest category in the GPT-Image2 prompt system?

While the exact distribution varies by commit, Photography & Realism and Illustration & Art typically contain the most individual prompt templates based on their line number ranges (lines 584-637 and 638-664 respectively). To determine the current largest category, run the Node.js aggregation script and count occurrences per category.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}