OpenAI launches ChatGPT Images 2.5 with faster generation and better edits

ChatGPT Images 2.5 is here with faster image generation and quality editing. This time, the new model introduced Sketch, a feature that lets users draw rough ideas directly in ChatGPT.OpenAI introduces ChatGPT Images 2.5. OneAI has introduced a new image generation model, ChatGPT Images 2.5, succeeding the current GPT-image-2. According to the company, the new model ships with a major focus on faster generation, better image quality and more accurate editing. For developers comes one of the biggest updates. GPT-Image-2.5 Flare, the default model for most API applications, offers higher-quality images than GPT-Image-2 while cutting latency by up to 50 per cent. In simpler terms, images should now take less time to generate without giving up on quality.In a blog post, OpenAI says people now create more than 3 billion images every week across ChatGPT Images and its GPT-Image models in the API. Now, with the new image-generation models, the AI firm aims to make images look more natural.Natural-looking imagesIt’s all about better lighting, texture and details that the new ones are designed to do. It can also hold on to the important features of people and objects when they are used as reference images.For example, if you upload a photo of a person and ask ChatGPT to place them in a completely different setting or visual style, Images 2.5 is better at keeping the person recognisable. More precise editingOne of the more useful improvements, according to the company, is the ability to make smaller changes without messing up everything else in the image.Want to change the background but keep the person untouched? Or replace a product while keeping the layout the same? Images 2.5 is designed to follow those instructions more closely. The model also gets better at handling multiple edits in the same conversation. So, if you keep asking for changes, earlier edits are more likely to stay intact instead of the image slowly turning into something you never asked for.You can now sketch your ideaExplanation in words is not an easy task for everyone. It can be harder. But there are other forms of explanation, one of which is simply drawing. And OpenAI include this form of explanation with a new Sketch feature in ChatGPT.Users can draw a rough idea directly in ChatGPT and use it as a visual guide for the final image. It could be a basic room layout, an outfit design or even a terrible doodle that somehow makes sense only to you.You can then add a text description explaining the style or other details. To try it, users can type “@Sketch” in ChatGPT.Two new models for developersOpenAI is also adding two image models to its API.GPT-Image-2.5 Flare is the default option for most applications. It brings the new quality and editing improvements while offering up to 50 per cent lower latency than GPT-Image-2.Then there is GPT-Image-2.5 Sunburst, which is aimed at more detailed creative work. It takes longer to generate images but is designed to offer tighter control over complex edits.For regular ChatGPT users, Images 2.5 is rolling out across desktop, mobile and web. The update essentially makes image creation quicker and editing more predictable, while giving users a new way to explain an idea when words simply aren't enough.- EndsPublished On: Sep 9, 2026 12:11 IST