After being overshadowed in the images vertical in recent months by Google’s Nano Banana and Nano Banana Pro, OpenAI has come up with a response.
OpenAI today unveiled ChatGPT Images, powered by its new flagship image generation model GPT Image 1.5, marking a significant upgrade in both creative capabilities and practical editing functions. The release comes at a critical moment for OpenAI, as Google’s Nano Banana and Nano Banana Pro image models have captured market attention in recent months. Those models went viral across social media, driving substantial new user growth to Gemini and establishing Google as a formidable competitor in AI-generated imagery. Now, OpenAI is fighting back with a model that it claims delivers more precise edits, faster generation, and better instruction following.
A Comprehensive Upgrade
GPT Image 1.5 introduces several key improvements over its predecessor. The model generates images up to four times faster than before while maintaining quality, addressing one of the main friction points users experienced with earlier versions. More importantly, it excels at preserving details during edits—a capability that directly addresses user frustrations with AI image editing tools that often produce inconsistent results.
The model’s editing capabilities span multiple approaches: adding, subtracting, combining, blending, and transposing elements within images. According to OpenAI, users can now make precise adjustments while maintaining consistent lighting, composition, and personal appearance across multiple iterations. This opens possibilities for practical applications like photo editing and virtual try-ons for clothing and hairstyles, alongside more creative transformations.
Text rendering has also seen notable advancement. GPT Image 1.5 can handle denser and smaller text more reliably than previous versions, an improvement that matters significantly for graphic design, marketing materials, and branded content creation.
A Dedicated Creation Space
Beyond the model improvements, OpenAI is introducing a dedicated Images feature within ChatGPT, accessible through the sidebar on mobile and web interfaces. This new space includes dozens of preset filters and prompts designed to jumpstart creative exploration, with regular updates to reflect emerging trends. The interface allows users to generate images without writing detailed prompts, lowering the barrier to entry for casual users.
This user experience focus represents OpenAI’s recognition that model capabilities alone don’t guarantee adoption—especially when competing against Google’s viral success with Nano Banana’s accessible, shareable outputs.
API Access and Enterprise Applications
For developers and businesses, GPT Image 1.5 is now available through OpenAI’s API at a 20% reduction in cost compared to GPT Image 1. The pricing change, combined with improved consistency in preserving branded elements across edits, positions the model as particularly attractive for marketing teams, e-commerce platforms, and creative tool developers.
OpenAI highlights the model’s ability to maintain branded logos and key visual elements across multiple edits, making it suitable for generating product image catalogs with various angles, scenes, and variants from a single source image. Early enterprise and startup adopters span creative tools, e-commerce, and marketing software sectors.
The Competitive Landscape
OpenAI’s release illustrates the intensifying competition in AI-generated imagery. Google’s Nano Banana models succeeded not just through technical capability but through viral adoption—they drew mainstream users to experiment with AI image generation. OpenAI’s response focuses on much of the same abilities such as image editing, with better editing, faster generation, and more reliable instruction following compared to previous versions.
The company acknowledges that while GPT Image 1.5 shows clear improvements over earlier versions in internal testing, results remain imperfect and “there is still significant room for improvement in future iterations.” This measured tone contrasts with the enthusiastic reception Google’s models received, perhaps suggesting OpenAI recognizes it’s playing catch-up in public perception even as it advances technical capabilities.
Whether ChatGPT Images can reclaim ground lost to Nano Banana Pro will depend not just on technical performance but on whether OpenAI can generate the same viral momentum that propelled Google’s offerings into mainstream consciousness. For now, the company is betting that superior editing precision and a streamlined user experience will be enough to bring image editing users back—and keep them there.