Today marks a significant leap forward for creative AI. OpenAI has just released a major upgrade to its ChatGPT Images feature, powered by a new flagship model that promises to transform how we generate and edit images. The core message is clear: whether you’re starting from a blank slate or refining an existing photo, the output will now more closely match what you envision in your mind.
The new model, available in ChatGPT for all users and in the API as GPT Image 1.5, brings a suite of powerful improvements designed to make image creation both more precise and more delightful.
Precision Editing That Respects Your Vision
One of the most impressive advancements is in image editing. The model now adheres to your intent with remarkable reliability, changing only what you ask for while preserving crucial elements like lighting, composition, and even a person’s appearance across multiple edits. This unlocks a new level of practical utility, enabling believable virtual try-ons for clothing and hairstyles, sophisticated photo edits, and stylistic transformations that retain the essence of the original.
For example, you can start with separate images of two men and a dog, combine them into a scene, add chaotic elements, transform individual styles, and even place them in a completely new setting—all while maintaining consistency.
Unleashing Creative Transformations
Beyond editing, the model’s creativity shines in its ability to reimagine concepts. It can change and add elements like text and layout to bring ideas to life—think turning a simple portrait into a vintage Hollywood movie poster—all while preserving important details. This is made even easier with a new dedicated Images experience in ChatGPT, featuring preset styles and prompts to jump-start inspiration without requiring a written prompt.
Sharper Instruction Following and Text Rendering
The model demonstrates significantly improved instruction following, enabling more intricate and precise compositions. A great test is creating a complex 6×6 grid of specific, varied objects—a task the new model handles with much greater accuracy than its predecessor.
Text rendering has also taken a substantial step forward. The model is now capable of handling denser, smaller text and complex layouts, such as accurately rendering a full newspaper article from markdown formatting.
A Faster, More Intuitive Workflow
Speed is a game-changer. The new model generates images up to 4 times faster, and you can continue queuing new ideas while others are still rendering. This, combined with the new dedicated creation space in the ChatGPT sidebar—complete with preset filters, trending prompts, and a one-time likeness upload feature—makes the creative process more fluid and exploratory than ever.
Real-World Improvements and Honest Limitations
OpenAI has been transparent about both the progress and the remaining challenges. When re-evaluating examples from the initial launch, the model shows clear improvements in areas like rendering complex scenes with many small faces and producing more natural-looking outputs.
However, limitations persist in handling multiple distinct faces in a single image, certain stylistic requests, and multilingual text generation. The team acknowledges there is significant room for future improvement.
Power for Developers and Businesses
For developers and enterprises, GPT Image 1.5 is available via the API with all the same enhancements. It offers stronger preservation of branded elements—making it ideal for marketing and e-commerce—and comes with the added benefit of being 20% cheaper for image inputs and outputs compared to the previous version. Industry leaders like Wix are already praising its high-fidelity results and reliability for production workflows.
Availability and Looking Ahead
The new ChatGPT Images model is rolling out now globally to all users. It works seamlessly across ChatGPT models, so no special selection is needed. The previous version will remain accessible as a custom GPT.
This release represents a meaningful step in the journey of AI-powered image generation. With its focus on precision, creativity, speed, and usability, it effectively puts a versatile creative studio in your pocket. The future promises even finer-grained control and richer outputs, but for now, the tools for bringing your visual ideas to life have just gotten a lot more powerful.