Image Generation

Stable

Generate high-quality images from text prompts using state-of-the-art AI models.

Generate Image

POST/api/v1/generate-image

Create images from text descriptions.

Request

cURL


curl -X POST "https://cloud.milady.ai/api/v1/generate-image" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "A serene mountain landscape at sunset",
    "model": "google/gemini-2.5-flash-image",
    "aspectRatio": "1:1",
    "numImages": 1
  }'

JavaScript


const response = await fetch('https://cloud.milady.ai/api/v1/generate-image', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    prompt: 'A serene mountain landscape at sunset',
    model: 'google/gemini-2.5-flash-image',
    aspectRatio: '1:1',
    numImages: 1,
  }),
});
 
const data = await response.json();
console.log(data.images[0].url);

Python


import requests
 
response = requests.post(
    'https://cloud.milady.ai/api/v1/generate-image',
    headers={
        'Authorization': 'Bearer YOUR_API_KEY',
        'Content-Type': 'application/json',
    },
    json={
        'prompt': 'A serene mountain landscape at sunset',
        'model': 'google/gemini-2.5-flash-image',
        'aspectRatio': '1:1',
        'numImages': 1,
    }
)
 
data = response.json()
print(data['images'][0]['url'])

Parameters

Parameter	Type	Required	Description
`prompt`	string	✓	Description of the image to generate
`model`	string		Model to use. Default: `google/gemini-2.5-flash-image`
`aspectRatio`	string		Output aspect ratio. Default: `1:1`
`numImages`	integer		Number of images (1-4). Default: 1
`stylePreset`	string		Style preset to apply (see below)
`sourceImage`	string		Base64 data URL for image-to-image generation

Available Aspect Ratios

Aspect Ratio	Description
`1:1`	Square (default)
`16:9`	Wide landscape
`9:16`	Tall portrait
`4:3`	Standard landscape
`3:4`	Standard portrait
`21:9`	Ultra-wide cinematic
`9:21`	Ultra-tall vertical

Style Presets

Preset	Description
`none`	No style modification (default)
`photographic`	Realistic lighting and details
`digital-art`	Vibrant colors, modern aesthetics
`comic-book`	Bold lines, dramatic shading
`fantasy-art`	Magical and ethereal elements
`analog-film`	Film grain and vintage tones
`neon-punk`	Cyberpunk style with neon colors
`isometric`	Isometric perspective
`low-poly`	Low-poly 3D style
`origami`	Paper-folding style
`line-art`	Clean line art with minimal shading
`cinematic`	Dramatic lighting and composition
`3d-model`	High-quality 3D rendered appearance

Response


{
  "images": [
    {
      "url": "https://your-storage.vercel-storage.com/images/abc123.webp",
      "text": "Optional text response from the model",
      "mimeType": "image/webp",
      "fileSize": 102400
    }
  ],
  "numImages": 1
}

The url field contains a permanent link to the generated image. The image field (base64) is only included as a fallback if blob storage upload fails.

Available Models

Model	Provider	Speed	Best For
`google/gemini-2.5-flash-image`	Google	Fast	General use (default)
`google/gemini-3.1-flash-image-preview`	Google	Fast	Preview/testing
`google/gemini-3-pro-image`	Google	Medium	Higher quality
`openai/gpt-5-nano`	OpenAI	Medium	OpenAI image generation
`bfl/flux-kontext-max`	BFL	Medium	Context-aware generation

Tip: Start with google/gemini-2.5-flash-image for quick iterations.

Image-to-Image Generation

Transform existing images by providing a sourceImage parameter:


const response = await fetch("https://cloud.milady.ai/api/v1/generate-image", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    prompt: "Transform this into a watercolor painting",
    sourceImage: "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAA...",
    model: "google/gemini-2.5-flash-image",
  }),
});

The sourceImage must be a base64-encoded data URL with the format data:{mimeType};base64,{base64Data}.

Prompt Tips

Be Specific

Good prompts include subject, style, lighting, and composition details:


{
  "prompt": "A golden retriever puppy playing in autumn leaves, golden hour lighting, shallow depth of field, professional photography"
}

Use Style Presets

Apply consistent styling using the stylePreset parameter:


{
  "prompt": "Portrait of a woman in natural lighting",
  "stylePreset": "photographic",
  "aspectRatio": "3:4"
}

Style Keywords

Enhance your prompts with these style modifiers:

Category	Keywords
Photography	`professional photo`, `DSLR`, `35mm film`, `macro`, `portrait`
Art Styles	`oil painting`, `watercolor`, `digital art`, `concept art`, `anime`
Lighting	`golden hour`, `studio lighting`, `dramatic shadows`, `soft light`
Quality	`highly detailed`, `8K`, `masterpiece`, `award winning`

Error Handling

Code	Error	Solution
400	Invalid prompt	Check prompt is non-empty string
400	Invalid model	Use a supported model from the list above
402	Insufficient credits	Add credits to your account
429	Rate limited	Wait and retry with exponential backoff
500	Generation failed	Retry, or try a different model

Example Error Response


{
  "error": "Insufficient credits for image generation",
  "required": 0.02
}

Batch Generation

Generate multiple images in a single request (up to 4):


curl -X POST "https://cloud.milady.ai/api/v1/generate-image" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Cyberpunk cityscape at night",
    "model": "google/gemini-2.5-flash-image",
    "numImages": 4,
    "aspectRatio": "16:9"
  }'

Each image in the batch is generated with the style preset applied.