Will GPT Image 2 Have Inpainting?

TL;DR

Probably yes, either explicitly or as part of a broader editing workflow. OpenAI already supports image editing in its current image stack, and inpainting is one of the most commercially useful editing patterns. That makes it a very likely GPT Image 2 capability. The uncertainty is branding and interface: OpenAI may present it as “editing,” “replace,” or “modify this area” instead of calling it inpainting directly.

Question	Best answer now
Is inpainting likely?	Yes
Confirmed for GPT Image 2?	No
Might it be renamed in product UI?	Yes

Why inpainting is such a strong bet

The current image generation guide already shows that OpenAI thinks beyond text-to-image. Once editing exists, regional editing is the obvious next expectation because users want to:

swap objects,
fix hands or faces,
change text areas,
and update backgrounds without regenerating the whole image.

Editing mode	Likelihood for GPT Image 2
Full image restyle	Very likely
Region-specific replace	Likely
Precision mask editing	Plausible, but interface unknown

My take

I would plan as if GPT Image 2 will include inpainting-level behavior, even if OpenAI describes it differently. That is especially true if the model is meant for design and commercial use cases rather than just casual image creation.

If you want the broader feature view, read Will GPT Image 2 Have Image Editing?. If your use case is localized text fixes, compare Will GPT Image 2 Support Chinese Text? and Will GPT Image 2 Support Japanese Text?.

Why this feature matters so much

Inpainting is where image models start to feel like design tools instead of novelty engines. Teams use it to fix a corner, swap an object, remove a logo, or change a label without losing the rest of the asset. If GPT Image 2 delivers that reliably, it will matter far more than a small jump in aesthetic quality.

So even without official confirmation, this is one of the most commercially credible forecasted features.

If you care about asset revision more than raw creation, inpainting may end up being the single most important GPT Image 2 capability to watch.

For many teams, it is the difference between “nice demo” and “actual production tool.”

FAQ

What is inpainting in simple terms?

It means changing part of an image without remaking the entire image from scratch.

Could GPT Image 2 support editing but not precision masks?

Yes. OpenAI might ship a simplified editing experience first, especially in ChatGPT.

Is inpainting more likely in the API or ChatGPT?

Probably both eventually, but the API may expose finer control sooner if OpenAI wants developers to build custom tools.

Sources

GPT Image Countdown is not affiliated with OpenAI. All trademarks belong to their respective owners.