Image Generation

Generate images from text descriptions using open-source image generation models including FLUX and SDXL.

Privacy-first: All image models are open-source and self-hosted on European infrastructure. Your prompts are never sent to external APIs.

POST /v1/images/generations

Quick Example

from openai import OpenAI

client = OpenAI(
    api_key="sk-mel-your-api-key-here",
    base_url="https://api.melious.ai/v1"
)

response = client.images.generate(
    model="flux-schnell",
    prompt="A serene mountain landscape at sunset with a reflective lake",
    size="1024x1024",
    n=1
)

print(response.data[0].url)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk-mel-your-api-key-here',
  baseURL: 'https://api.melious.ai/v1'
});

const response = await client.images.generate({
  model: 'flux-schnell',
  prompt: 'A serene mountain landscape at sunset with a reflective lake',
  size: '1024x1024',
  n: 1
});

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

curl https://api.melious.ai/v1/images/generations \
  -H "Authorization: Bearer sk-mel-your-api-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux-schnell",
    "prompt": "A serene mountain landscape at sunset with a reflective lake",
    "size": "1024x1024",
    "n": 1
  }'

Request Parameters

Parameter	Type	Required	Default	Description
`prompt`	string	Yes	-	Text description of the image to generate
`model`	string	No	`flux-schnell`	Model ID (see Available Models)
`n`	integer	No	1	Number of images (1-4)
`size`	string	No	`1024x1024`	Image dimensions
`response_format`	string	No	`url`	`url` or `b64_json`
`user`	string	No	null	End-user identifier

Available Models

Model	Brand	Description	Best For
`flux-schnell`	Black Forest Labs	Fast generation	Quick iterations, prototyping
`flux-dev`	Black Forest Labs	High-quality generation	Production images
`sdxl-base-1.0`	Stability AI	SDXL base model	General purpose
`sdxl-base-v10`	Stability AI	SDXL base v10	General purpose
`sdxl-lightning-4step`	ByteDance	Fast 4-step generation	Speed priority
`sdxl-lightning-8step`	ByteDance	Quality 8-step generation	Quality priority

Available Sizes

Size	Aspect Ratio	Description
`512x512`	1:1	Square (small)
`768x768`	1:1	Square (medium)
`1024x1024`	1:1	Square (large)
`1024x768`	4:3	Landscape
`768x1024`	3:4	Portrait
`1280x720`	16:9	Widescreen
`720x1280`	9:16	Vertical

Available sizes may vary by model. Check the model documentation for specific size support.

Response Format

{
  "created": 1699999999,
  "data": [
    {
      "url": "https://..../generated-image.png"
    }
  ],
  "environment_impact": {
    "carbon_g_co2": 2.5,
    "water_liters": 0.01,
    "energy_kwh": 0.005,
    "renewable_percent": 85,
    "provider_id": "nebius",
    "location": "NL"
  }
}

Field	Type	Description
`created`	integer	Unix timestamp
`data`	array	Array of generated images
`data[].url`	string	Image URL (valid for 1 hour)
`data[].b64_json`	string	Base64 image data (if requested)

Base64 Response

Request base64-encoded images instead of URLs:

response = client.images.generate(
    model="flux-schnell",
    prompt="A futuristic cityscape",
    size="1024x1024",
    response_format="b64_json"
)

import base64
image_data = base64.b64decode(response.data[0].b64_json)
with open("image.png", "wb") as f:
    f.write(image_data)

Prompt Guidelines

Descriptive and Specific:

A cozy coffee shop interior with warm lighting, wooden furniture,
plants on shelves, rain visible through large windows, soft jazz
atmosphere, watercolor painting style

With Style Reference:

A portrait of a cat wearing a crown, oil painting style,
Renaissance era, dramatic lighting, rich colors

Technical Specifications:

Product photography of a sleek smartphone, white background,
soft shadows, 45-degree angle, professional studio lighting

Be specific - Include details about style, lighting, colors
Specify medium - "oil painting", "3D render", "photograph"
Describe composition - "close-up", "wide shot", "aerial view"
Set mood - "serene", "dramatic", "whimsical"
Avoid ambiguity - "a dog" vs "a golden retriever puppy"

Error Handling

Error Code	Description	Solution
`BILLING_INSUFFICIENT_ENERGY`	Not enough balance	Top up credits
`VALIDATION_INVALID_VALUE`	Invalid size or parameter	Check allowed values
`INFERENCE_PROVIDER_ERROR`	Generation failed	Retry or modify prompt

Content Policy: Prompts that violate content policies will be rejected. Avoid explicit, violent, or harmful content.

Image Generation

Image Generation

Quick Example

Request Parameters

Available Models

Available Sizes

Response Format

Base64 Response

Prompt Guidelines

Error Handling

See Also

On this page