← All models

gpt-image-2 ★

OpenAIImage
0.11 Credits
Get your API key

Playground

Configure a request and see the result here.

Input

Mode
Estimated maximumUnavailable

Final credits follow successful outputs and any account discount. This estimate does not reserve your balance.

Result

Ready

No image generated yet

API

Endpoints, runnable examples and the parameters used by this model.

Endpoints

POST/openai/images/generations

Send your API key as a Bearer token. Keep it on the server in production.

cURL
curl -X POST 'https://api.qiyaov.com/openai/images/generations' \
-H 'authorization: Bearer <YOUR_API_KEY>' \
-H 'content-type: application/json' \
-d '{
  "model": "gpt-image-2",
  "prompt": "",
  "size": "1024x1024",
  "n": 1,
  "async": true
}'

Core parameters

modelgpt-image-2
promptInstructions for the generated image
size / n / asyncCanvas, output count and asynchronous mode
Full API documentation ↗

Pricing

Credits are the billing unit. Dollar amounts depend on the selected package.

Packages

We offer various credit packages. The more you purchase, the higher the discount you receive.

up to46% OFF
Top-up

You pay$13,584for 142,800 credits · $0.0951 / credits46% OFF

Route pricing

Routegpt-image-2
Billing rule0.11 Credits per successful image
Routegpt-image-2:reverse
Billing rule0.11 Credits per successful image
Routegpt-image-2:official
Billing ruleBilled by actual token usage

The official route is usage metered; its final price is known after the response. An account discount may change the charge.

Request calculator

Current Playground requestUnavailable
Images1
Canvas size1024x1024
ModeGenerate

Final credits follow successful outputs and any account discount. This estimate does not reserve your balance.

Overview

Capabilities and limits that matter before integration.

What this model is good at

Text-rich creative

Posters and layouts with words in the composition.

Structured prompts

Scenes with explicit count, placement and visual hierarchy.

Flexible canvas

Common sizes and valid custom dimensions.

Model facts

InputText; an image and optional mask for editing
OutputImage returned by an asynchronous task
Images1–10
Custom sizeMultiples of 16; longest edge at most 3840 pixels