OpenAI's new image model edits with surgical precision

The biggest headache with AI photo editing has always been the butterfly effect: ask the model to change a jacket, and it suddenly gives your subject a new face and different lighting.

OpenAI just rolled out a major overhaul to ChatGPT Images, powered by its new flagship image generation model, GPT Image 1.5. The goal this time isn't just prettier pixels—it is all about precision and speed.

Precise edits, 4x faster

The most noticeable upgrade is speed. Image generation is now up to four times faster than before, making the back-and-forth iteration cycle significantly less tedious.

More importantly, editing has become remarkably surgical. When you upload an image and ask for changes, the model actually preserves elements like lighting, composition, and personal likeness. It handles adding, removing, blending, and transposing elements without rewriting the rest of the canvas.

This makes practical tasks—like virtual clothing try-ons or swapping hairstyles—substantially more reliable. The model also handles smaller, denser text rendering and renders tiny faces in crowded scenes far more naturally.

OpenAI also introduced a dedicated Images home in ChatGPT's sidebar across web and mobile. Instead of forcing you to engineer complex prompts, it features preset styles and visual filters you can apply with a click.

The update is live globally for ChatGPT users and in the API as GPT Image 1.5, with Enterprise and Business access rolling out later. API costs for inputs and outputs are also 20% cheaper compared to GPT Image 1.

Why it matters

We think this launch marks a crucial pivot from random visual generation to reliable, repeatable production.

Until now, creative professionals and e-commerce teams have struggled to weave diffusion models into production pipelines simply because local edits were unpredictable. A simple client request like "keep the model, change the bag" usually broke the entire render.

By prioritizing visual consistency, logo fidelity, and cheaper API pricing, OpenAI is directly targeting commercial utility. E-commerce merchants can now realistically generate an entire catalog of product variants and contextual scenes from a single master shot without watching their budget vanish.

It is not flawless yet—OpenAI explicitly notes that outputs can still miss the mark on complex spatial setups. But by pairing surgical consistency with a dedicated UI inside ChatGPT, OpenAI is turning image generation from a party trick into an everyday utility.