OpenAI has introduced an upgraded version of its image-generation and editing tool, now called ChatGPT Images, enhancing competition with Google’s Nano Banana image editor. The update features GPT Image 1.5, which OpenAI describes as its new flagship image model, available immediately within ChatGPT and through the company’s API.
According to OpenAI, the new model offers faster image generation, up to four times quicker than its predecessor, along with more precise edits and improved consistency across outputs. This upgrade targets users seeking reliable image creation, whether starting from scratch or modifying existing photos.
In addition to the model update, OpenAI has added a dedicated Images tab in the ChatGPT app and web interface. This new section serves as a workspace for generating images and a discovery feed to inspire creative ideas and experimentation.
While OpenAI positions this update as a significant refinement rather than a complete overhaul, early benchmarks indicate strong performance. GPT Image 1.5 has quickly risen to the top of the LMArena Text-to-Image leaderboard, surpassing Google’s Nano Banana Pro. However, initial real-world comparisons suggest that results may vary by task, with some users reporting comparable or slightly better outcomes from Google’s tool in certain scenarios.
The new ChatGPT Images feature is being rolled out globally and should be accessible to users immediately, although free-tier access remains limited. Developers can also integrate GPT Image 1.5 via the OpenAI API, with pricing details available on OpenAI’s website.
—
