Skip to main content
The OpenAI GPT-Image series is one of the most advanced AI image generation model families, supporting text-to-image generation, image editing, and more. APIYI fully supports the GPT-Image series with complete compatibility with the OpenAI official Image API format.

​
Model Comparison

FeatureGPT-Image-1.5GPT-Image-1GPT-Image-1-Mini
Release DateDecember 2025April 2025April 2025
Generation SpeedUltra fast (under 10s)StandardFaster
Text RenderingExcellent (dense text, tables)GoodGood
Editing CapabilityHigh-precision element preservationStandardStandard
Instruction FollowingMore preciseStandardStandard
Input Price$5.00 / million tokens$2.50 / million tokensLower
Output Price$10.00 / million tokens$8.00 / million tokensLower
Best ForProfessional design, branding, infographicsGeneral image generationRapid prototyping, batch generation
How to Choose?
  • Need precise text rendering, brand assets, infographics → GPT-Image-1.5
  • General image generation, budget-conscious → GPT-Image-1
  • Rapid iteration, batch production → GPT-Image-1-Mini

​
Quick Start

All GPT-Image models share the same API format — just change the model parameter to switch: 
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.apiyi.com/v1"
)

{/* Using GPT-Image-1.5 */}
response = client.images.generate(
    model="gpt-image-1.5",
    prompt="A professional product showcase with white background and soft lighting",
    size="1024x1024",
    quality="high"
)

print(response.data[0].url)

​
Common Parameters

All GPT-Image models support the following parameters:
ParameterTypeDescription
modelstringgpt-image-1.5, gpt-image-1, or gpt-image-1-mini
promptstringImage description (max 1000 characters)
sizestring1024x1024, 1536x1024, 1024x1536, auto
qualitystringlow, medium, high, auto
output_formatstringpng, jpeg, webp
backgroundstringtransparent, opaque, auto
nintegerNumber of images (1-10)

​
Detailed Documentation

GPT-Image-1.5

Latest flagship model, 4x speed boost, superior text rendering, LMArena #1

GPT-Image-1

Classic image generation model, ideal for general use, cost-effective

​
FAQ

GPT-Image-1.5 is 4x faster, ~20% cheaper, with significantly improved text rendering (supporting dense text and tables), and more precise editing that preserves original elements. GPT-Image-1 is best for general image generation at a lower price point.
Yes, images generated via the API are fully yours to use, including for commercial purposes.
Yes, but English prompts generally produce the best results. For scenes requiring Chinese text rendering, GPT-Image-1.5 has significantly stronger text capabilities.
Simply change the model parameter from gpt-image-1 to gpt-image-1.5. All other parameters are fully compatible — no code changes needed.

Sora Image Generation

Reverse-engineered image generation, $0.01/image ultra-low price

Nano Banana Pro

4K HD, best-in-class text rendering, $0.05/image