GPT Image 2

OpenAI's latest image generation model, supporting high-quality text-to-image, transparent backgrounds, and flexible image compression options.

Try it Online

Test in Playground ↗ · View Models ↗

Quick Start

from openai import OpenAI

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

response = client.images.generate(
    model="gpt-image-2",
    prompt="A golden retriever on a beach at sunset, photorealistic",
    size="1024x1024",
    quality="high",
    n=1
)

print(response.data[0].url)

Endpoint

Item	Value
Method	POST
Path	/v1/images/generations
Base URL	https://open.dieyuyun.com
Protocol	OpenAI Images API

Authentication

Bearer Token: Authorization: Bearer YOUR_API_KEY

Standard Request Fields

Field	Type	Required	Default	Description
model	string	Yes	—	Model identifier: gpt-image-2
prompt	string	Yes	—	Image description text, up to 32,000 characters
size	string	No	1024x1024	Image size: 1024x1024, 1536x1024 (landscape), 1024x1536 (portrait), auto
quality	string	No	auto	Generation quality: auto, low, medium, high
n	integer	No	1	Number of images to generate, 1–10
response_format	string	No	url	Response format: url or b64_json
background	string	No	auto	Background type: auto, opaque, transparent
moderation	string	No	auto	Content moderation level: auto, low
output_compression	integer	No	—	Output image compression level, 0–100, only effective with b64_json
output_format	string	No	—	Output format: png, jpeg, webp

Model-Specific Parameters

Transparent Background (background)

Set background to transparent to generate PNG images with a transparent background, ideal for logos, stickers, and overlays.

{
  "model": "gpt-image-2",
  "prompt": "A company logo of a mountain",
  "background": "transparent",
  "output_format": "png"
}

Output Compression (output_compression)

When response_format is b64_json, use output_compression to control the compression level of the output image, reducing payload size.

{
  "model": "gpt-image-2",
  "prompt": "A landscape painting",
  "response_format": "b64_json",
  "output_compression": 80
}

Request Examples

cURL

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A golden retriever on a beach at sunset",
    "size": "1024x1024",
    "quality": "high",
    "n": 1
  }'

Python

from openai import OpenAI

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

response = client.images.generate(
    model="gpt-image-2",
    prompt="A golden retriever on a beach at sunset",
    size="1024x1024",
    quality="high",
    n=1
)

print(response.data[0].url)

Node.js

import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://open.dieyuyun.com/v1',
  apiKey: 'YOUR_API_KEY',
})

const response = await client.images.generate({
  model: 'gpt-image-2',
  prompt: 'A golden retriever on a beach at sunset',
  size: '1024x1024',
  quality: 'high',
  n: 1,
})

console.log(response.data[0].url)

Transparent Background Example

cURL

curl https://open.dieyuyun.com/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A minimalist logo of a rocket",
    "size": "1024x1024",
    "background": "transparent",
    "output_format": "png",
    "n": 1
  }'

Python

from openai import OpenAI

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

response = client.images.generate(
    model="gpt-image-2",
    prompt="A minimalist logo of a rocket",
    size="1024x1024",
    background="transparent",
    output_format="png",
    n=1
)

print(response.data[0].url)

Response Format

Success Response

{
  "created": 1717000000,
  "data": [
    {
      "url": "https://...",
      "revised_prompt": "A photorealistic image of a golden retriever running on a sandy beach during golden hour sunset, warm light reflecting off the ocean waves"
    }
  ]
}

Field	Type	Description
created	integer	Unix timestamp
data	array	Array of generated results
data[].url	string	Image URL (when response_format is url)
data[].b64_json	string	Base64-encoded image data (when response_format is b64_json)
data[].revised_prompt	string	The revised/optimized prompt used for generation

Error Response

See Error Codes for details.

Best Practices

• Be specific in prompts: GPT Image 2 responds better to detailed descriptions. Include style, lighting, composition, and other details to significantly improve generation quality.
• Use transparent backgrounds: When designing logos, icons, or assets that need to be composited, set background: "transparent" to skip post-processing.
• Choose the right quality: Use low for rapid prototyping and iteration, high for final output. Use auto to let the system choose based on prompt complexity.
• Pick the right size: Use 1536x1024 for landscape content, 1024x1536 for portrait content, and 1024x1024 for square content.
• Batch generation: Use the n parameter to generate multiple images at once for comparison, which is more efficient than making separate requests.

Rate Limits

See Rate Limits for details.

Related Documentation

• GPT Image 2 Official - Use the official native API
• Model List & Pricing - Browse available models and pricing
• Playground - Test image generation interactively