OpenAI ChatGPT Images | GPT Image 1.5 improves visual editing
OpenAI has released a new version of ChatGPT Images, powered by GPT Image 1.5, its latest flagship image generation model. Published on December 16, 2025, the update improves precise editing, detail preservation, instruction following, text rendering, and generation speed, while adding a dedicated Images experience inside ChatGPT for faster creative exploration.
ChatGPT Images gets faster generation and more precise visual editing
OpenAI’s update focuses on making image generation more useful for real creative workflows. The new model is designed to follow editing instructions more reliably while keeping important visual details intact, including lighting, composition, people’s appearance, and other elements that often need to remain consistent across revisions.
For designers, the key change is control. GPT Image 1.5 is not only about creating images from scratch; it is also meant for iterative editing, visual transformations, branded graphics, ecommerce assets, marketing concepts, and production work where small changes need to preserve the structure of the original image.
How GPT Image 1.5 improves ChatGPT Images
OpenAI says the new model is better at making edits to uploaded images without changing unrelated details. It can add, remove, combine, blend, and transpose elements while preserving the parts of an image that should remain stable, making it more practical for photo edits, try-ons, style changes, and concept refinements.
The update also improves instruction following and text rendering. This matters for designers because image models often struggle with dense text, small labels, structured layouts, and multi-step visual requirements. Better text handling makes the tool more useful for posters, infographics, mockups, presentation visuals, and brand-oriented graphics.
New creative options for designers and visual teams
The new Images experience in ChatGPT gives users a dedicated creation space in the sidebar, with preset filters and prompts designed to make visual exploration faster. That can help creators test visual directions without writing every prompt from scratch, while still using natural language for more specific edits.
For production teams, GPT Image 1.5 is especially relevant because the API version supports the same improvements as ChatGPT Images. OpenAI says image inputs and outputs are now 20% cheaper compared with GPT Image 1, while stronger preservation of logos and key visuals makes the model more useful for marketing, ecommerce, and brand workflows.
The model still has limitations, and OpenAI notes that results remain imperfect. For professional use, designers should continue reviewing composition, text accuracy, brand consistency, likeness handling, accessibility, licensing, and final export quality before using generated images in public or client-facing work.
Availability and API access
OpenAI says the new ChatGPT Images model is rolling out globally to all ChatGPT users and API users. It works across models, so users do not need to manually select a separate model in order to use the new image generation experience.
GPT Image 1.5 is also available in the API, giving developers and creative platforms access to the same improvements for image generation, editing, preservation, text rendering, and faster iteration. The earlier ChatGPT Images version remains available as a custom GPT for users who still need it.
Sources and Recommended Links
- The new ChatGPT Images is here | OpenAI (Official)
- ChatGPT | OpenAI (Official)
- OpenAI API Platform | OpenAI (Official)
