Image generation

Generate images from text prompts using DALL·E 3 or FLUX.2-pro. Return base64-encoded JSON or a hosted URL.

Endpoint

POSThttps://api.getnimbus.net/v1/images/generations

Request body

ParameterTypeRequiredDescription
modelstringYesdall-e-3 or FLUX.2-pro
promptstringYesText description of the desired image. Max 4000 chars.
nintegerNoNumber of images to generate. Default 1, max 4.
sizestringNo256x256, 512x512, 1024x1024 (default), 1792x1024, 1024x1792.
response_formatstringNourl (default) or b64_json.

Example: cURL

curl https://api.getnimbus.net/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $MERIDIAN_API_KEY" \
  -d '{
    "model": "dall-e-3",
    "prompt": "A cyberpunk cat wearing a violet hoodie, neon city background",
    "n": 1,
    "size": "1024x1024",
    "response_format": "b64_json"
  }'

Example: Python

import requests

headers = {
    "Authorization": f"Bearer {os.environ['MERIDIAN_API_KEY']}",
    "Content-Type": "application/json",
}

payload = {
    "model": "FLUX.2-pro",
    "prompt": "A cyberpunk cat wearing a violet hoodie, neon city background",
    "n": 1,
    "size": "1024x1024",
    "response_format": "url",
}

r = requests.post(
    "https://api.getnimbus.net/v1/images/generations",
    headers=headers,
    json=payload,
)

print(r.json())

Response

When response_format is b64_json:

{
  "created": 1717027200,
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAA...",
      "revised_prompt": "A cyberpunk cat wearing a violet hoodie..."
    }
  ]
}

When response_format is url (default):

{
  "created": 1717027200,
  "data": [
    {
      "url": "https://cdn.getnimbus.net/gen/abc123.png",
      "revised_prompt": "A cyberpunk cat wearing a violet hoodie..."
    }
  ]
}

Rate limits

Free tier: 5 images / hour. Pro tier: 100 images / hour. Enterprise: custom. Headers x-ratelimit-remaining and x-ratelimit-reset are included on every response.

← Back to docsChat completions →