Skip to content
OpenDream AI

Model

GPT Image 2 AI Image Generator

GPT Image 2 is an OpenAI image model available for both text-to-image and image-to-image work. It accepts up to 16 input images, renders at 1K, 2K or 4K, and can produce a transparent background at 1K. That makes it a practical pick for assets that need to drop straight into a layout.

What GPT Image 2 does well

Best for

Prompt examples

Frequently asked questions

Can GPT Image 2 make images with a transparent background?
Yes, but only at 1K resolution. The transparent option is not available at 2K or 4K.
Which aspect ratios work at higher resolutions?
At 2K, the 5:4 and 4:5 ratios are unavailable, and square 1:1 images go up to 2K. The studio only offers combinations the model supports.
How many reference images can I use with GPT Image 2?
Image-to-image requests accept up to 16 input images.

Related models

Try GPT Image 2

Or let Smart mode choose the best model for each request.

Start creating

Capabilities in OpenDream

Model capabilities
CapabilityGPT Image 2
Text to imageYes
Edit / image to imageYes
Reference imagesUp to 16
Resolutions1K · 2K · 4K
Aspect ratios10
Negative promptNo
SeedNo
Transparent backgroundYes
SpeedMedium
QualityVery high
Credits per image (from)6

Live from the OpenDream model registry. Costs can vary with resolution and reference images; the exact cost is shown before you generate.