In the latest release of OpenAI, the image generation offering has become more precise and consistent update ChatGPT imagesas more and more businesses and brands utilize AI image generation to lend a hand visualize designs.
The updates will be rolled out to all ChatGPT users and API as a GPT 1.5 image. The company claims to be powered by GPT 5.2, which many early adopters have found to be a powerful upgrade for business utilize.
“Many people’s first experience with ChatGPT is turning a text prompt into an image,” said Fidji Simo, OpenAI’s general manager for applications, in the Substack post. “It’s a magical way to see what this technology can do, but the chat interface wasn’t originally designed for this. Creating and editing images is a different kind of task and deserves a space built with visuals in mind.”
Business-friendly updates in precise editing and following instructions
One of the biggest updates to ChatGPT images is more targeted editing, even if the image is generated in a chat platform rather than via an API. Image generation models such as ChatGPT Images, Google Nano Banana, and Stable Diffusion tout suggestion-based enhancements to AI-created photos, where the user can point to specific parts of the photo to change. But these features can sometimes be hit and miss.
With the update, OpenAI said the model better adapts to user expectations “while maintaining consistency of elements such as lighting, composition, and people’s appearance across inputs, outputs, and subsequent edits.”
Users can instruct the model to perform most types of image editing, such as adding or subtracting an element, combining, blending, and transposing.
OpenAI said this model “follows instructions more reliably” than previous versions. It is also able to render text better and generate actual, readable letters, even if they are denser or smaller. OpenAI has updated the model to create better, smaller faces in photos of immense groups of people.
“These transformations work for both simple and more complex concepts and are easy to try out using pre-built styles and ideas in the new ChatGPT Images feature – no written prompt required,” says OpenAI.
Battle of image generators
The OpenAI image model update follows Google’s update the highly praised Nano Banana Pro an image model that has gained recognition in the programming community.
The company must compete with other constantly evolving, constantly improving image generation models aimed at attracting more enterprise users. OpenAI has to face more than just Google. Alibaba announced this in August this image of Qwen can render readable text in both Chinese and English. Black Forest Labs released Stream.2which also offers a tough open source image model.