Generate images from text prompts — fully OpenAI Images Generations API compatible (DALL·E 2/3, gpt-image-1, etc.)

Create images from a text prompt. Fully compatible with the OpenAI Images Generations API — supports dall-e-2, dall-e-3, gpt-image-1, and similar models.

Field	Value
Method	`POST`
Path	`/v1/images/generations`
Full URL	`https://api.amux.ai/v1/images/generations`
Auth	`Authorization: Bearer <API_TOKEN>`
Request format	`application/json`
Response format	`application/json`

This endpoint is fully aligned with the OpenAI Images Generations protocol. You can use the official OpenAI SDK directly — just override the Base URL and API Key.

Request Body

Prop

Type

Passthrough: amux-api forwards any field not listed above to the upstream as-is. Vendor-specific fields such as watermark / watermark_enabled (Zhipu / Aliyun Tongyi), user_id (Aliyun), and extra_body (channel-private params used by Gemini Nano Banana for aspect_ratio / image_size / seed) are recognized and routed to the appropriate channels.

Examples

curl https://api.amux.ai/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AMUX_API_KEY" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "A cute baby sea otter wearing a beret, watercolor style",
    "n": 1,
    "size": "1024x1024",
    "quality": "high",
    "background": "transparent",
    "output_format": "png"
  }'

import os
import base64
from openai import OpenAI

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

response = client.images.generate(
    model="gpt-image-1",
    prompt="A cute baby sea otter wearing a beret, watercolor style",
    n=1,
    size="1024x1024",
    quality="high",
)

image_b64 = response.data[0].b64_json
with open("otter.png", "wb") as f:
    f.write(base64.b64decode(image_b64))

import { writeFileSync } from 'node:fs';
import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://api.amux.ai/v1',
  apiKey: process.env.AMUX_API_KEY,
});

const response = await client.images.generate({
  model: 'gpt-image-1',
  prompt: 'A cute baby sea otter wearing a beret, watercolor style',
  n: 1,
  size: '1024x1024',
  quality: 'high',
});

const b64 = response.data![0].b64_json!;
writeFileSync('otter.png', Buffer.from(b64, 'base64'));

Response Example

{
  "created": 1730000000,
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAA...",
      "revised_prompt": "A watercolor illustration of a cute baby sea otter wearing a beret..."
    }
  ],
  "usage": {
    "total_tokens": 1234,
    "input_tokens": 28,
    "output_tokens": 1206,
    "input_tokens_details": {
      "text_tokens": 28,
      "image_tokens": 0
    }
  }
}

{
  "error": {
    "message": "...",
    "type": "invalid_request_error",
    "param": "prompt",
    "code": "invalid_value"
  }
}

HTTP Status	Meaning
`200`	Success
`400`	Bad request parameters (missing `prompt`, unsupported `size`, etc.)
`401`	Missing or invalid API key
`403`	Model not authorized or blocked by content safety policy
`404`	Model not found
`429`	Rate limit triggered or insufficient balance
`500` / `502` / `503`	Server-side error or upstream unavailable

Generate Images

Request Body

Examples

Response Example

Response Fields

ImageData Fields

Error Response

On this page