OpenAI continues on the ‘code red’ warpath with a new graphics model

OpenAI is developing a The new version of ChatGPT Graphics that promises better tracking, more accurate editing, and 4x faster speed.

The new version, called GPT Image 1.5, is available from Tuesday to all ChatGPT users and through the API. It’s the latest escalation in competition with Google’s Gemini after OpenAI CEO Sam Altman last month announced a “code red” in a leaked internal memo. The memo detailed OpenAI’s plans to regain its position as the AI ​​leader after Google began to gain market share by following suit. Gemini 3 releaseits latest version, is Nano Banana Pro, the latest version of Google’s viral image generator – both top-rated. LMArena leaderboard on many benchmarks.

Google maintains its lead even after OpenAI responded well last week with implementation of GPT-5.2and rank it as its highest quality construction and professional work every day. OpenAI is said to be planning to release a new generator in early January, accelerating those plans with this week’s announcement. Its last image release was GPT Image 1 in April.

GPT 1.5 is where graphic and video designers go beyond prototypes and get ready for production. Like Nano Banana ProChatGPT Image offers post-production features, providing flexible controls to keep things consistent, such as face shape, lighting, texture, and color tone across all edits.

Most GenAI graphics tools are bad at replication, so this can be a big part. When asked to make certain changes, such as “changing the shape of the face” or “making the light cooler,” models often interpret the entire image, resulting in inconsistencies.

The change isn’t just about innovation. ChatGPT images can also be accessed through a sidebar of ChatGPT that works “like a production studio,” Fidji Simo, CEO of OpenAI, he wrote in a blog post on Tuesday.

“Viewing and editing new images makes it easy to create images that match your vision or are inspired by current trends and filters,” Simo wrote.

On top of the new graphics generator, OpenAI is introducing new ways to update ChatGPT events with visual effects. The plan is to enable search queries to show better visuals with clearer sources, which could be useful for tasks such as changing measurements or watching sports games, according to Simo.

“When you’re creating, you need to see and create what you’re creating. When images tell a story better than words alone, ChatGPT should include it,” Simo wrote. “When you want to respond quickly or take action when you’re in another device, it needs to be there. When we do this, we can close the gap between what’s in your mind and your ability to bring it to life.”