AI Art Generation

turnip uses OpenAI's GPT Image 1.5 - the most advanced image generation model available. Learn how to craft effective prompts and leverage advanced features.

GPT Image 1.5 Overview

GPT Image 1.5 is OpenAI's state-of-the-art image generation model, offering superior instruction following, text rendering, and detailed editing capabilities.

Prompt Engineering

Crafting effective prompts is key to generating great token artwork. Here are some best practices:

Be Specific and Descriptive

# Less effective
"A cat"

# More effective
"A hyper-realistic 3D render of a fluffy orange tabby cat 
wearing AR glasses, sitting on a pile of gold coins, 
cyberpunk neon lighting, detailed fur texture"

Style Keywords

Include style keywords to guide the aesthetic:

Token-Specific Prompts

For meme tokens and crypto art, consider including:

Advanced Features

Image Streaming

Watch your image generate in real-time with streaming preview:

$ turnip generate "Epic dragon breathing fire" --streamStreaming generation... Partial 1/3: [████░░░░░░] 33% Partial 2/3: [████████░░] 67% Partial 3/3: [██████████] 100% ✓ Final image saved!

Image Editing (Inpainting)

Edit existing images with prompts:

$ turnip edit ./my-image.png --prompt "Add a golden crown on the cat's head"

Multi-Image Reference

Combine multiple reference images:

$ turnip generate "Combine these elements into one artwork" \ --ref ./logo.png \ --ref ./character.png \ --fidelity high

Cost Optimization

Image generation costs scale with quality and size. Here's a token breakdown: