The new ChatGPT Images is here [1.5] (simonwillison.net)

OpenAI has launched an upgraded version of its ChatGPT Images feature, now labeled as GPT Image 1.5, which enhances speed and instruction adherence. This model processes image inputs and outputs up to four times faster and reduces costs by approximately 20% compared to its predecessor, GPT Image 1. The update aims to improve user experience, particularly for those needing precise edits and detailed image generation, which is crucial in a competitive arena where rivals like Google's Nano Banana are also evolving rapidly. The significance of this release lies in its potential to attract even more users and solidify OpenAI’s position in the AI/ML landscape. The new capabilities are particularly evident when generating text-heavy graphics, which now rival those produced by competitive models. Early tests demonstrate that GPT Image 1.5 can produce distinctive and engaging imagery, such as adding whimsical elements to photos, while also delivering improved functionality for infographics. These advancements mark an important step forward in image generation technology, aligning with the growing demand for faster and more cost-effective AI solutions in creative applications.