On this page

• Basic request
• Response format
• Choosing a model

Image Generation Guide

Generate a new image from a text prompt. This is different from PHP Image Editing, which edits an existing uploaded image — this endpoint creates one from scratch.

POST https://api.oblion.io/v1/images/generations

Basic request

curl https://api.oblion.io/v1/images/generations \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "gpt-image-1",
    "prompt": "A watercolor painting of a fox in an autumn forest",
    "n": 1,
    "size": "1024x1024"
  }'

from openai import OpenAI

client = OpenAI(base_url="https://api.oblion.io/v1", api_key="YOUR_API_KEY")

response = client.images.generate(
    model="gpt-image-1",
    prompt="A watercolor painting of a fox in an autumn forest",
    n=1,
    size="1024x1024",
)
print(response.data[0].url)

import OpenAI from "openai";

const openai = new OpenAI({ baseURL: "https://api.oblion.io/v1", apiKey: "YOUR_API_KEY" });

const response = await openai.images.generate({
  model: "gpt-image-1",
  prompt: "A watercolor painting of a fox in an autumn forest",
  n: 1,
  size: "1024x1024",
});
console.log(response.data[0].url);

Response format

By default the API returns a temporary url you can download the image from. Some models also support response_format: "b64_json" to get the image inline as base64 instead:

response = client.images.generate(
    model="gpt-image-1",
    prompt="A minimalist logo of a mountain",
    response_format="b64_json",
)
import base64
image_bytes = base64.b64decode(response.data[0].b64_json)
with open("logo.png", "wb") as f:
    f.write(image_bytes)

Choosing a model

See the full list under Image Generation & Editing in Supported Models. A few notable options:

Some image models are hosted with limited capacity. If a request returns 429 ("upstream load saturated"), wait a few seconds and retry — this is expected under peak load, not a bug in your request.