How do I create image variations from one prompt through an API?

To create multiple image variations from a single prompt via an API, you generally do one of these:

1. Generate multiple images in one request by setting a n / count parameter.
2. Generate one image and vary it using an image editing/variation endpoint, if the API supports it.
3. Reuse the same prompt with different seeds/styles to get controlled diversity.

If you mean OpenAI’s image API specifically, the simplest approach is usually multiple images from one prompt.

Example: generate multiple images from one prompt

Python

from openai import OpenAI

client = OpenAI()

result = client.images.generate(
    model="gpt-image-1",
    prompt="A futuristic city skyline at sunset, cinematic lighting, highly detailed",
    size="1024x1024",
    n=4
)

for i, img in enumerate(result.data):
    print(f"Image {i+1}: {img.url}")

cURL

curl https://api.openai.com/v1/images/generations \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "A futuristic city skyline at sunset, cinematic lighting, highly detailed",
    "size": "1024x1024",
    "n": 4
  }'

If you want more distinct variations

Try changing:

• style words: “photorealistic”, “anime”, “oil painting”, “minimalist”
• camera/composition: “wide angle”, “close-up”, “top-down”
• mood/lighting: “dramatic lighting”, “soft natural light”
• seed: if supported by your provider, use different seeds for reproducible variation

If you want variations of an existing image

Use an edit/variation endpoint with an input image, if available in your API. That’s better when you want to preserve composition while changing details.

Good practice for prompts

A strong prompt often includes:

• subject
• environment
• style
• lighting
• composition
• quality/detail hints

Example:

“A red sports car parked on a rainy Tokyo street at night, neon reflections, cinematic, ultra-detailed, shallow depth of field”

If you want, I can show you:

• a Node.js example
• a Python batch script
• or how to do image variations with seeds and prompt templates