The image-generating feature of ChatGPT that OpenAI said brought its servers to their knees last month is expanding to other apps and generative AI developers, the company announced Wednesday.
The ability to make images directly in ChatGPT caused a huge spike in usage and spawned trends like people turning themselves into action figures. OpenAI said more than 130 million users worldwide created more than 700 million images in the first week after the feature launched.
Now OpenAI is releasing the image model, dubbed gpt-image-1, into its API, meaning developers can add the image-generating feature to their products and tools.
Those developers include Adobe, which will provide access to the model in its Firefly and Express apps, according to OpenAI’s blog post. Other design tools using the model include Figma, HeyGen and Wix.
Read also: AI Essentials: 27 Ways to Make Gen AI Work for You, According to Our Experts
Microsoft, meanwhile, announced that OpenAI’s GPT-4o image generation feature would now be available in its Microsoft 365 Copilot app through a new “Create” experience. The tool can make images from prompts, turn PowerPoint presentations into videos and more.
On the text side, OpenAI rolled out a host of new models last week. That includes GPT-4.1, new models available to developers that can handle more information and work more quickly, and OpenAI o3, reasoning models that promise improved performance in coding, math and visual understanding.
Watch this: Everything Announced at Google Cloud Next
11:36
This recent run of announcements by OpenAI is the latest move in a race with Google for the lead in the gen AI market. Google’s Gemini has its own image-generation features with a model called Imagen 3, improvements to which were announced earlier this month at Google Cloud Next 2025.
