ChatGPT Images is now faster and better than ever | Infinium-tech
Following the recent release of GPT-5.2, today OpenAI is unveiling new and improved ChatGPT Images, which is four times faster at generating images. It has apparently been developed to compete with Google’s Nano Banana Pro, in the same way that GPT-5.2 is a direct answer to the Gemini 3 Pro.
OpenAI’s press release states that ChatGPT images “follow your intent more reliably” when you ask it for editing, “down to the smallest details”. It changes only what you ask for, while leaving everything else intact in input, output, and subsequent edits.
This should unlock “results that match your intent”, allowing ChatGPT to “act as a creative studio in your pocket, capable of both practical editing and expressive reimagining”. The new model is better at adding, subtracting, combining, blending and moving, “so you can get the transformation you want without losing what makes the image special”.
inside the new ChatGPT Images Feature You’ll get preset styles and ideas that you can choose from without having to provide written prompts. The model follows instructions more reliably, enabling more precise editing and more complex original compositions, “where relationships between elements are preserved as desired”. Text rendering should also be improved, as well as the rendering of many small faces.
The new ChatGPT Images are now rolling out to all users globally.
Leave a Reply