Image Generation

Overview

The image generation endpoint creates images from text descriptions. It follows the OpenAI images API format and supports multiple image generation models across providers.

POST https://api.universal-ai.dev/v1/images/generations

Request

Headers

Header	Required	Description
`Authorization`	Yes	`Bearer YOUR_API_KEY`
`Content-Type`	Yes	`application/json`

Body Parameters

Parameter	Type	Required	Description
`model`	string	No	The image model to use (e.g., `dall-e-3`, `cf/stable-diffusion-xl`). Defaults to `dall-e-3`.
`prompt`	string	Yes	A text description of the image to generate. Be specific and descriptive for best results.
`n`	integer	No	Number of images to generate. Default: `1`. Max depends on the model.
`size`	string	No	Image dimensions. Options vary by model. Common values: `1024x1024`, `1024x1792`, `1792x1024`. Default: `1024x1024`.
`quality`	string	No	Image quality. `standard` or `hd` (for supported models). Default: `standard`.
`response_format`	string	No	Format of the returned image: `url` (default) or `b64_json`.
`style`	string	No	Image style: `natural` or `vivid` (DALL-E 3 only). Default: `vivid`.

Example Request

curl https://api.universal-ai.dev/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "dall-e-3",
    "prompt": "A serene mountain lake at sunset with pine trees reflected in still water, photorealistic style",
    "n": 1,
    "size": "1024x1024",
    "quality": "hd"
  }'

Response

{
  "created": 1700000000,
  "data": [
    {
      "url": "https://api.universal-ai.dev/v1/files/img_abc123.png",
      "revised_prompt": "A serene mountain lake at sunset, with tall pine trees perfectly reflected in the crystal-clear, still water. The sky is painted in warm hues of orange, pink, and purple."
    }
  ]
}

Response Fields

Field	Type	Description
`created`	integer	Unix timestamp of when the image was generated.
`data`	array	Array of generated image objects.
`data[].url`	string	URL of the generated image (when `response_format` is `url`).
`data[].b64_json`	string	Base64-encoded image data (when `response_format` is `b64_json`).
`data[].revised_prompt`	string	The prompt as revised by the model (DALL-E 3 may refine your prompt).

SDK Examples

Python:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.universal-ai.dev/v1"
)

response = client.images.generate(
    model="dall-e-3",
    prompt="A futuristic city skyline with flying cars and neon lights",
    n=1,
    size="1024x1024"
)

image_url = response.data[0].url
print(f"Image URL: {image_url}")

JavaScript / TypeScript:

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://api.universal-ai.dev/v1",
});

const response = await client.images.generate({
  model: "dall-e-3",
  prompt: "A futuristic city skyline with flying cars and neon lights",
  n: 1,
  size: "1024x1024",
});

console.log(`Image URL: ${response.data[0].url}`);

Available Models

Model ID	Provider	Sizes	Description
`dall-e-3`	OpenAI	1024x1024, 1024x1792, 1792x1024	High-quality image generation with prompt understanding
`dall-e-2`	OpenAI	256x256, 512x512, 1024x1024	Faster generation, multiple images per request
`cf/stable-diffusion-xl`	Cloudflare	1024x1024	Stable Diffusion XL on Workers AI (low cost)
`cf/stable-diffusion-v1-5`	Cloudflare	512x512	Stable Diffusion 1.5 on Workers AI
`flux-bfl/flux-1.1-pro`	Black Forest Labs	1024x1024, custom	FLUX 1.1 Pro for high-fidelity generation

Tips for Better Results

Be specific — describe the subject, style, lighting, and composition. "A golden retriever sitting in a sunny meadow, oil painting style" produces better results than "a dog."
Specify style — include artistic direction like "photorealistic," "watercolor," "digital art," or "pencil sketch."
Use quality: "hd" — for DALL-E 3, HD mode produces images with finer detail and greater consistency.
Try different models — each model has different strengths. FLUX excels at photorealism, DALL-E 3 at creative interpretation, and Stable Diffusion at stylized art.

On this page