OpenAI image model on Crazyrouter
All OpenAI models →

Try GPT Image 1.5 online — generate an image, then call the API

GPT Image 1.5 is available on Crazyrouter through the OpenAI-compatible images endpoint. Pay per image, no subscription, same key as your text models.

Use this model via API

This model is available through the paid API. Create an account, add funds and create an API key. The $2 Playground credit is for supported text models only, not API calls.

Sign up for API access
Image generationText to imageImage editing
Get API KeyView live pricing

Paste the model ID, live price, base URL and a working request straight into your coding agent.

OpenAI
gpt-image-1.5
$0.165
Per image
Model ID
gpt-image-1.5
Provider
OpenAI
Endpoint
/v1/images/generations
Task
image generation
Input modalities
Text, image
Output modalities
Image
Release date
2025-10-01

What is GPT Image 1.5?

GPT Image 1.5 is a GPT Image image model from OpenAI for text-to-image generation and image editing. On Crazyrouter it shares an API key and balance with chat, video and audio models, so image generation slots into an existing integration without a second vendor account.

OpenAI images schema

Same request and response shape as the OpenAI Images API; existing client code keeps working.

Per-image billing

You are charged per generated image at the rate on this page, nothing else.

Reference images

Where the upstream model supports it, pass reference images for editing and style transfer.

One key, all modalities

Mix image generation with chat, video and music models under one balance.

GPT Image 1.5 API quickstart

POST to https://api.crazyrouter.com/v1/images/generations with model gpt-image-1.5. The response follows the OpenAI images schema (URL or base64).

cURL
curl https://api.crazyrouter.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $CRAZYROUTER_API_KEY" \
  -d '{
    "model": "gpt-image-1.5",
    "prompt": "Product photo of a matte ceramic coffee mug on a light oak desk, soft morning window light, shallow depth of field, neutral background, commercial photography",
    "n": 1,
    "size": "1536x1024"
  }'

GPT Image 1.5 sample outputs

Every sample below was generated through Crazyrouter with the request shown above, unretouched and uncurated; latency and cost are what that single run actually took.

Product photo
Product photoLatency 55.9s · This run $0.107
Show full prompt

Product photo of a matte ceramic coffee mug on a light oak desk, soft morning window light, shallow depth of field, neutral background, commercial photography

Studio portrait
Studio portraitThis run $0.107
Show full prompt

Studio portrait of a smiling woman in her thirties wearing a linen shirt, natural window light, 85mm lens look, soft grey backdrop, photorealistic

Night street, aerial
Night street, aerialThis run $0.107
Show full prompt

Top-down aerial view of a rain-soaked city street at night, neon signs reflecting on wet asphalt, a single yellow umbrella, cinematic, moody

Typography poster
Typography posterThis run $0.107
Show full prompt

Minimal poster with the word CRAZYROUTER in bold black sans-serif letters centered on a warm off-white background, one thin amber underline, lots of negative space, print quality

Generated 2026-09-28

Best use cases

Product and marketing visuals at scale
Blog and social cover images
Design exploration and mood boards
Image editing, inpainting and style transfer
Game and app asset prototyping
Personalised visuals in user-facing apps

Works with developer tools

OpenAI SDKs

client.images.generate(model="gpt-image-1.5", prompt=...) against the Crazyrouter base URL.

n8n / Dify / Make

Use the HTTP node or the OpenAI-compatible connector with the Crazyrouter endpoint.

Custom backends

Plain HTTPS; no SDK required.

Crazyrouter playground

Try prompts in the console before writing code.

GPT Image 1.5 FAQ

Which endpoint do I call?

https://api.crazyrouter.com/v1/images/generations with model set to gpt-image-1.5.

How is it billed?

Per generated image at the price shown on this page; failed generations are not charged.

Which sizes and formats are supported?

The upstream model's sizes and formats; pass size and response_format as documented by the vendor.

Can I edit an existing image?

If the upstream model supports edits, send reference images via the images edits endpoint or the image field.

Is there a rate limit?

Limits follow your account tier; contact support for higher throughput.