OpenAI continues its ‘Code Red’ war path with new image generation model

OpenAI is rolling out a new version of ChatGPT Images that promises better command following, more precise editing, and up to 4x faster image generation speeds.

This new model, called GPT Image 1.5, will be available to all ChatGPT users via the API starting Tuesday. This is the latest escalation in Google’s competition with Gemini after OpenAI CEO Sam Altman declared a “code red” in a leaked internal memo last month. The memo details OpenAI’s plans to regain its position as an AI leader after Google began gaining market share following the release of its latest flagship model, Gemini 3, and Nano Banana Pro, the latest version of Google’s viral image generator, both of which topped the LMArena leaderboard in multiple benchmarks.

Google remains in the lead after OpenAI responded to its success with the release of GPT-5.2 last week, touting it as its most advanced model yet for developers and everyday professional use. OpenAI was reportedly planning to release a new image generator in early January, and this week’s announcement accelerated those plans. The last image model release was GPT Image 1 in April.

GPT Image 1.5 arrives as image and video generators advance beyond prototypes and gain more production-ready capabilities. Similar to Nano Banana Pro, ChatGPT Image offers post-production capabilities and provides more detailed editing controls to maintain visual consistency such as facial similarity, lighting, composition, and tone throughout the edit.

This is a huge improvement since most GenAI image tools are not good at iterating. When asked to make a specific change, such as adjusting facial expressions or making the lighting cooler, the model often reinterprets the entire image, leading to a lack of consistency.

Updates aren’t just about new features. ChatGPT images can also now be accessed through a dedicated entry point in the ChatGPT sidebar, which functions “more like a creative studio,” Fidji Simo, CEO of applications at OpenAI, said in a blog post on Tuesday.

“New image viewing and editing screens make it easy to create images that match your vision and get inspired by trending prompts and preset filters,” Simo wrote.

In addition to a new image generator, OpenAI is introducing new ways to improve your ChatGPT experience with more visual elements. According to Simo, the plan is to surface more visuals with clear sources in search queries, which could be useful for tasks like converting measurements or checking sports scores.

“When you’re creating, you need to be able to see and articulate what you’re creating. If visuals tell a story better than words alone, ChatGPT should include it,” Simo wrote. “If you need an answer right away, or if your next step is in another tool, it should be right there. That way, we can continue to close the distance between what’s in your head and your ability to make it happen.”