POST /v1/images/generations — Generate Images from Text

The /v1/images/generations endpoint generates images from natural-language prompts using state-of-the-art diffusion models such as FLUX, all served from Meliai’s European provider network. The API is OpenAI-compatible, so you can use the official OpenAI SDK’s images.generate method by pointing it at Meliai’s base URL. Generated images are returned as either a URL or base64-encoded data, and every response includes the standard environment_impact and billing_cost fields.

Endpoint

POST https://api.meliai.ai/v1/images/generations

Authorization: Bearer sk-mel-<KEY> via Authorization header.

Parameters

string

required

The image generation model ID. FLUX-family models and other open-weight image models available on the Meliai network are listed under GET /v1/models. Routing flavor suffixes such as :price are supported.

string

required

A natural-language description of the image you want to generate. More specific, detailed prompts generally produce higher-quality results.

integer

Number of images to generate in a single request. Default: 1. Higher values multiply credit usage proportionally.

string

Dimensions of the output image in "WxH" format, e.g. "1024x1024", "1280x720". Available sizes depend on the model. Defaults to the model’s native resolution if omitted.

string

How the generated image is returned:

"url" — a temporary URL pointing to the image (default)
"b64_json" — the image encoded as a base64 string inside the JSON response

Examples

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["MELIAI_API_KEY"],
    base_url="https://api.meliai.ai/v1",
)

response = client.images.generate(
    model="<IMAGE_MODEL_ID>",
    prompt="A panoramic view of the Rhine valley at sunset",
    n=1,
    size="1024x1024",
)
print(response.data[0].url)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.MELIAI_API_KEY,
  baseURL: "https://api.meliai.ai/v1",
});

const response = await client.images.generate({
  model: "<IMAGE_MODEL_ID>",
  prompt: "A panoramic view of the Rhine valley at sunset",
  n: 1,
  size: "1024x1024",
});
console.log(response.data[0].url);

curl https://api.meliai.ai/v1/images/generations \
  -H "Authorization: Bearer $MELIAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<IMAGE_MODEL_ID>",
    "prompt": "A panoramic view of the Rhine valley at sunset",
    "n": 1,
    "size": "1024x1024"
  }'

Retrieve as base64

To embed the image directly in your application without a follow-up HTTP request, set response_format to "b64_json":

import base64

response = client.images.generate(
    model="<IMAGE_MODEL_ID>",
    prompt="A medieval European castle surrounded by autumn forests",
    response_format="b64_json",
)

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

curl https://api.meliai.ai/v1/images/generations \
  -H "Authorization: Bearer $MELIAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<IMAGE_MODEL_ID>",
    "prompt": "A medieval European castle surrounded by autumn forests",
    "response_format": "b64_json"
  }'

Response

{
  "created": 1720000000,
  "data": [
    {
      "url": "https://cdn.meliai.ai/images/generated/abc123.png",
      "b64_json": null
    }
  ],
  "environment_impact": {
    "energy_kwh": 0.00048,
    "carbon_g_co2": 0.034,
    "water_liters": 0.0019,
    "renewable_percent": 89,
    "pue": 1.22,
    "provider_id": "provider-eu-south",
    "location": "Paris, FR"
  },
  "billing_cost": {
    "energy": 0.00024,
    "credits": 0.0038,
    "paid_with": "credits"
  }
}

integer

Unix timestamp of when the image was generated.

array

Array of image objects, one per generated image. Each contains:

url — temporary download URL (present when response_format is "url")
b64_json — base64-encoded image data (present when response_format is "b64_json")

object

Per-request environmental footprint. Fields: energy_kwh, carbon_g_co2, water_liters, renewable_percent, pue, provider_id, location.

object

Itemised cost: energy (EUR), credits deducted, and paid_with.

Image URLs are temporary and expire after a short period. Download and store any images you intend to keep, or use response_format: "b64_json" to receive the image data directly.

Image generation runs entirely on European infrastructure. Your prompts and generated images are never used to train models and never leave the EU.

Overview

Inference

Batch & Files

Models

POST /v1/images/generations — Generate Images from Text

Endpoint

Parameters

Examples

Retrieve as base64

Response

​Endpoint

​Parameters

​Examples

​Retrieve as base64

​Response

Endpoint

Parameters

Examples

Retrieve as base64

Response