‹Explore/openai/gpt-image-2

OpenAI

GPT Image 2

"vendor": "openai""model_name": "gpt-image-2"

OpenAI's next-generation image model supporting text-to-image generation and image editing up to 4K resolution with multiple quality tiers

Text To ImageImage To ImageNewFeatured

auto_storiesLLMs

descriptionSchema menu_bookModel Docs

VENDOROpenAI

INPUTtitleimage

OUTPUTimage

PRICING

Hover to view▾

Playground Setup

Beta

Analyzing model...

Output

Beta

Waiting for Input

README for GPT Image 2

Overview

GPT Image 2 is OpenAI''s next-generation image model, supporting both text-to-image generation and image editing via a unified task-based API. It delivers photorealistic outputs up to 4K resolution with four quality tiers, batch generation of up to 4 images, and PNG / JPEG / WebP output formats.

Key Features

Text-to-Image generation - Create images from natural language prompts
Image editing - Modify existing images with text instructions; optional mask for region-targeted edits
Multi-resolution output - From 1024x1024 up to 4K (3840x2160) with aspect ratio variants
Quality tiers - high, medium, low, and auto to balance quality vs. speed
Batch generation - Up to 4 images per request
Output formats - PNG (default), JPEG, WebP

Size Options

Size	Aspect Ratio
`1024x1024`	1:1 Square
`1536x1024`	3:2 Landscape
`1024x1536`	2:3 Portrait
`2048x2048`	1:1 Large
`2048x1152`	16:9 Widescreen
`3840x2160`	4K Landscape
`2160x3840`	4K Portrait
`auto`	Model chooses (default)

Pricing

Token-Based Billing (per 1M tokens)

Text tokens:
Input: $5.00
Cached input: $1.25
Output: $10.00

Image tokens:
Input: $8.00
Cached input: $2.00
Output: $30.00

Quality affects cost significantly (default: high). Token cost is ceiled to the nearest cent.

Parameters

nintegerDefault: 1

Number of edited images to generate.

maskstring

Optional mask image (URL or Base64-encoded) to specify the edit region.

sizeEnumDefault: "auto"

Output image resolution.

Possible values: 1024x1024, 1536x1024, 1024x1536, 2048x2048, 2048x1152, 3840x2160, 2160x3840, auto

formatEnumDefault: "png"

Output image format.

Possible values: png, jpeg, webp

imageslist<string>* required

Input images to edit. Each item is a URL or Base64-encoded string.

promptstring* required

Text prompt describing the desired edit.

GPT Image 2

Pricing Card

Playground Setup

Output

README for GPT Image 2

README for GPT Image 2

Overview

Key Features

Size Options

Pricing

Parameters