Image Generation

POST /v1/images/generations

Generate one or more images from a text prompt. Compatible (with extensions) with the OpenAI Images API.

Headers

Request body

Examples

curl -X POST https://api.quantized.us/v1/images/generations \
  -H "Authorization: Bearer sk-quantized-YOUR-KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1.5",
    "prompt": "A whimsical watercolor of a sleepy hedgehog under a mushroom",
    "size": "1024x1024",
    "quality": "low"
  }'

curl -X POST https://api.quantized.us/v1/images/generations \
  -H "Authorization: Bearer sk-quantized-YOUR-KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1.5",
    "prompt": "A logo on a fully transparent background, single hedgehog character",
    "quality": "medium",
    "background": "transparent",
    "output_format": "png"
  }'

curl -X POST https://api.quantized.us/v1/images/generations \
  -H "Authorization: Bearer sk-quantized-YOUR-KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.1-flash-image",
    "prompt": "A whimsical watercolor of a sleepy hedgehog under a mushroom"
  }'

import base64
from openai import OpenAI

client = OpenAI(
    api_key="sk-quantized-YOUR-KEY",
    base_url="https://api.quantized.us/v1",
)

resp = client.images.generate(
    model="gpt-image-1.5",
    prompt="A whimsical watercolor of a sleepy hedgehog",
    size="1024x1024",
    quality="low",
)
img_bytes = base64.b64decode(resp.data[0].b64_json)
with open("hedgehog.png", "wb") as f:
    f.write(img_bytes)

import base64
import httpx

response = httpx.post(
    "https://api.quantized.us/v1/images/generations",
    headers={"Authorization": "Bearer sk-quantized-YOUR-KEY"},
    json={
        "model": "gpt-image-1.5",
        "prompt": "An overhead view of an open notebook with a pen",
        "size": "1024x1024",
        "quality": "low",
    },
    timeout=180,
)
data = response.json()
img_bytes = base64.b64decode(data["data"][0]["b64_json"])
print(f"image: {len(img_bytes)} bytes, watermark: {data['data'][0]['watermark']}")

Response

{
  "created": 1787941342,
  "model": "gpt-image-1.5",
  "data": [
    {
      "b64_json": "iVBORw0KGgoAAAANSUhEUgAA...",
      "revised_prompt": null,
      "seed": null,
      "watermark": "c2pa",
      "flagged": false
    }
  ],
  "usage": {
    "images": 1,
    "input_tokens": 16,
    "output_tokens": 272,
    "total_tokens": 288,
    "megapixels": null,
    "credits_used": 1790500
  }
}

Response fields

Watermarking

Different providers/models embed different provenance metadata. The watermark enum in each data[] entry lets clients disclose this to end users if needed.

Per-model constraints

Each model has stricter limits than the serializer’s generic [1, 10] range for n and unrestricted size. The router does not enforce these — they’re handled by upstream:

The gpt-image-* models accept background, output_format, output_compression and moderation, and require OpenAI org verification on our account. The Gemini Flash Image models route through :generateContent and return mixed text and image parts. Imagen routes through :predict.

gpt-image-1 is still resolvable but is deprecated in the catalog. DALL-E 2 and DALL-E 3 were retired by OpenAI in 2026 and are no longer served.

Discovery via /v1/models

Filter GET /v1/models on output_modality.image == true:

models = response.json()["data"]  # GET /v1/models returns {"object": "list", "data": [...]}
image_models = [m for m in models if m["output_modality"]["image"]]

The catalog exposes per-model pricing in cost, whose shape varies by model (per_image for Imagen, prompt / completion for the token-priced models, and {} where the adapter computes the rate itself).

Providers

These are the only two providers holding the image_generation capability, so X-Quantized-Provider: bedrock is not usable here. Routing is resolved per-model from the catalog, so requesting gemini-3.1-flash-image reaches Gemini with no header needed.

Errors

reached Gemini  404  models/<id> is not found for API version v1beta ...
reached OpenAI  400  The model '<id>' does not exist.