OpenAI released an upgraded version of ChatGPT Images, powered by a new flagship image generation model designed to deliver more accurate edits while preserving visual details. The update allows users to create and modify images with greater control, producing results up to four times faster than the previous version.
The model improves precision when editing uploaded images, changing only the elements specified in prompts while maintaining consistency in lighting, composition, and human features. OpenAI said this enables more reliable photo edits, clothing and hairstyle try-ons, stylistic filters, and conceptual transformations without degrading the original image’s integrity.
The new model excels across multiple editing modes, including adding, removing, blending, and transposing elements. It also improves instruction-following, enabling more complex compositions where spatial and visual relationships remain intact. Text rendering has also advanced, allowing the model to handle denser and smaller typography more clearly.
Additional quality improvements include better handling of images with many faces and more natural-looking outputs, making generated images more immediately usable for creative and professional applications.
New Images Workspace and Creative Tools
Alongside the model upgrade, OpenAI introduced a dedicated Images space within ChatGPT, accessible via the sidebar on mobile and on chatgpt.com. The workspace is designed to streamline creative exploration, offering preset styles, filters, and ideas that allow users to generate images without writing detailed prompts.
The Images feature is rolling out globally to all ChatGPT users and is also available through the API as GPT Image 1.5. The earlier version of ChatGPT Images will remain accessible as a custom GPT.
OpenAI said the new model still has limitations, noting that while performance improved across many internal evaluations, results remain imperfect. The company framed the release as a step forward rather than a final solution, with further refinements planned in future iterations.
GPT Image 1.5 Expands API Capabilities
GPT Image 1.5 brings the same improvements to developers using the API, with stronger image preservation and more consistent handling of branded assets such as logos and key visuals. OpenAI positioned the model as well suited for marketing, ecommerce, and creative software use cases, including generating full product image catalogs from a single source image.
The company also reduced image input and output costs by 20% compared with GPT Image 1, allowing teams to iterate more within the same budget. Enterprises and startups across industries, including ecommerce and marketing software, are already using the updated model.
Quality Focus Amid Competitive Pressure
The release aligns with OpenAI’s broader emphasis on improving product quality and reliability. Earlier this year, the company paused plans to introduce advertising in ChatGPT after declaring an internal “code red,” shifting focus toward performance, stability, and security as competition in AI accelerates. The upgraded ChatGPT Images model reflects that priority, emphasizing accuracy, usability, and consistency over rapid feature expansion.
OpenAI said the new ChatGPT Images experience is available immediately across all surfaces and models, with more enhancements planned as image generation capabilities continue to evolve.