OpenAI Unveils New Image Generator Amid Intense AI Race
▼ Summary
– OpenAI has released GPT-Image-1.5, a new ChatGPT image model with better instruction-following, precise editing, and up to 4x faster generation.
– This release is part of OpenAI’s competitive response to Google’s Gemini and Nano Banana Pro models, which have taken market share and topped leaderboards.
– The model introduces advanced post-production features, allowing granular edits to maintain visual consistency like facial likeness and lighting across changes.
– ChatGPT Images is now accessible via a dedicated sidebar entry point designed to function more like a creative studio for users.
– OpenAI plans to enhance ChatGPT with more visual elements in search results, aiming to close the gap between a user’s idea and its creation.
OpenAI has launched a significantly upgraded version of its ChatGPT image generation tool, delivering major improvements in speed, precision, and user control. This new model, GPT-Image-1.5, is now available to all ChatGPT users and via its API, marking a strategic move in the intensifying competition with rivals like Google. The release follows an internal “code red” declared by OpenAI’s leadership, a response to Google gaining market traction with its own advanced models, Gemini 3 and the viral image generator Nano Banana Pro. Both Google offerings have recently led performance benchmarks, putting pressure on OpenAI to innovate rapidly.
The introduction of GPT-Image-1.5 represents a focused effort to reclaim leadership in the AI space. It arrives just after OpenAI’s release of the GPT-5.2 model for developers, with the image generator’s launch accelerated from earlier plans. A key advancement is its dramatically faster processing, promising image generation speeds up to four times quicker than its predecessor. Beyond raw speed, the model excels at following detailed user instructions and enables more precise editing, allowing for adjustments to specific elements like facial expression or lighting without altering the entire composition.
This capability to iterate on an image while maintaining visual consistency, such as preserving a person’s likeness or a scene’s color tone, addresses a common weakness in many generative AI tools. Typically, asking a model for a minor change can result in a completely reinterpreted image. OpenAI’s new system aims to provide granular control that feels more like professional post-production software. To facilitate this creative process, ChatGPT Images now has a dedicated entry point in the ChatGPT interface, designed to function as an integrated creative studio.
According to Fidji Simo, OpenAI’s CEO of Applications, the updated viewing and editing screens are built to help users materialize their vision more easily. The platform also offers inspiration through trending prompts and preset filters. Alongside the new image generator, OpenAI is enhancing the broader ChatGPT experience by integrating more visual elements into search responses. The goal is to provide clear, sourced visuals for queries where images are more informative than text alone, such as checking sports scores or converting measurements.
This holistic push toward a more visual and iterative creative tool reflects a broader industry trend where AI image and video generators are evolving from novel prototypes into practical, production-ready applications. By closing the gap between a user’s idea and its realization, OpenAI is positioning its tools as essential for both everyday creativity and professional use.
(Source: TechCrunch)
