GPT Image API pricing. Up to 62% below list.

GPT Image is OpenAI's family of image generation and editing models. GPT Image 2.5 Sunburst is the most capable, tuned for quality and precision. GPT Image 2.5 Flare is the fastest for high-quality, everyday images, and GPT Image 2 is built for production visuals such as infographics, posters and text-heavy layouts. On Zurelay every image is made at high quality: GPT Image 2 costs one flat price at any size, the 2.5 models are priced by size (1K, 2K, 4K), and each model edits and combines up to 16 reference images.

GPT Image models
3
From, per image
$0.02
Below OpenAI's price
47–62%

Pricing

Every GPT Image model. Live prices, one key.

Each model on Zurelay, with OpenAI's own price beside ours.

GPT Image API prices on Zurelay
ModelMakesPriceCompared withYou save
GPT Image 2.5 Flaregpt-image-2.5-flareAny size, high quality
from $0.028
per image
$0.0527
OpenAI, 1K
47%off
GPT Image 2.5 Sunburstgpt-image-2.5-sunburstAny size, high quality
from $0.026
per image
$0.0527
OpenAI, 1K
51%off
GPT Image 2gpt-image-2Any size, high quality
$0.02
per image
$0.053
OpenAI, 1K
62%off

Prices in US dollars, live from the catalog. Images are compared with the lab's own price per image at the smallest size. Each model's own page has every price it charges.

Examples

Made with GPT Image 2.5 Sunburst. Straight from the API.

Every image here is a single API call with the prompt underneath it, not retouched or upscaled. They're shown as compressed WebP; the API returns full-quality PNG.

Make your own
  • Portrait of a young woman with red hair and freckles sitting by a large window, soft morning light falling across her face, a knitted cream sweater, green eyes in sharp focus, natural skin texture, shot on medium format film, 80mm lens, photorealistic editorial portrait.
  • Golden hour in the Icelandic highlands: a lone red wooden cabin beside a braided glacial river, low mist drifting over black volcanic sand, snow-capped mountains lit pink by the setting sun, ultra-detailed landscape photograph, 24mm lens, polarizing filter.
  • Sunlit Scandinavian living room, a linen sofa in oatmeal tones, a light oak floor, a ceramic vase with dried grasses, floor-to-ceiling windows looking onto a pine forest, soft shadows, architectural digest interior photography, photorealistic.
  • Close-up of espresso pouring from a bottomless portafilter into a glass cup, rich tiger-striped crema, a thin golden stream, steam, a dark café background with warm bokeh, 100mm macro, high shutter speed, commercial coffee photography.
  • Interior of an old independent bookshop, tall wooden shelves packed with books, a rolling library ladder, a chalkboard sign on the counter that reads "Staff Picks This Week", warm lamp light, dust in a sunbeam, 35mm photograph, photorealistic.
  • A ruby-throated hummingbird hovering beside a red trumpet flower, wings frozen mid-beat, iridescent green feathers in sharp detail, soft green garden background, wildlife photography, 600mm lens, 1/8000s shutter.

Quickstart

One request. OpenAI's format.

Point the OpenAI SDK at https://api.zurelay.com/v1 and call images.generate, or images.edit with your reference images. Sizes, aspect ratios and every option are in the image guide.

Add credit from $10, with no subscription. The quickstart has the rest, and every GPT Image model has its own page.

generate.py
from openai import OpenAI
client = OpenAI(
base_url="https://api.zurelay.com/v1",
api_key="YOUR_ZURELAY_KEY",
)
image = client.images.generate(
model="gpt-image-2.5-sunburst",
prompt="A red fox sitting in fresh snow at dawn, photo",
size="1536x1024",
quality="high",
)
print(image.data[0].url)

Verified

The GPT Image you asked for. Tested before it serves.

Model capacity is often bought ahead, through prepaid credits, committed-spend contracts and volume tiers, and not all of it gets used. Sellers offer it below the usual price. Zurelay tests each route, pools several per model and sends every request to one that is available and verified.

Every image that comes back is measured against the size and aspect ratio you asked for. One at the wrong size, or an error sent as a result, counts as a failed attempt: the request moves to another route and you aren't billed for it. How models are verified.

Live health for every model is on the status page.

FAQ

GPT Image API questions.

Anything else? Ask our team and we’ll answer by email.

How much does the GPT Image API cost?

On Zurelay: GPT Image 2.5 Flare from $0.028 per image; GPT Image 2.5 Sunburst from $0.026 per image; GPT Image 2 $0.02 per image. That's 47–62% below OpenAI's own price. There's no subscription or minimum spend: you add prepaid credit from $10 and pay for what you use.

Is it the real GPT Image?

Yes. Each model is served over several routes run by independent sellers, and every route is tested before it serves requests and re-tested while it does. Images are measured against the size and aspect ratio you asked for, and one at the wrong size counts as a failed attempt you aren't billed for. If a result doesn't look right, send its x-request-id to support and we check the route that served it.

How do I start calling GPT Image through Zurelay?

Create a key, then call images.generate in the OpenAI SDK, or POST /v1/images/generations, at https://api.zurelay.com/v1 with a model ID such as gpt-image-2.5-sunburst. Code that already uses OpenAI's images API only needs the base URL, the key and the model name changed.

Does a bigger image cost more?

Not on GPT Image 2: on Zurelay it has one price per image at any size, always at high quality. GPT Image 2.5 Flare and Sunburst are priced by size: up to 1024x1024 is 1K, up to 2560x1440 is 2K, and larger is 4K. OpenAI bills every model by output tokens, so its larger images cost more.

Does it work with the OpenAI SDK?

Yes. Call images.generate or images.edit with the model ID, with the SDK's base URL set to https://api.zurelay.com/v1 and your Zurelay key.

Stop paying list price.
Start saving today.

Create a key in seconds, point the SDK you already use at Zurelay, and every request costs up to 90% less from the first token.