OpenAI’s updated 4o Image Generation image generator, part of ChatGPT, has attracted a wide audience thanks to its ability to create Studio Ghibli-style images. Now the company has opened up access to its “inherently multimodal model” via API — and major partners have already started booking access to the gpt-image-1 platform.
Image source: openai.com
«The model’s versatility allows it to create images in a variety of styles, accurately respond to user queries, apply knowledge about the world, and accurately render text, opening up countless practical application scenarios across a variety of domains,” OpenAI said in a blog post.
Users of Adobe and Figma services will be able to work with the model, having already included support for it in their toolkits. Adobe has added it to the Firefly and Express applications; from today, the gpt-image-1 image generator is also available in Figma Design, where it can adjust styles, add or remove objects, expand background images, and perform many other operations using simple text queries.
OpenAI added that it is continuing to collaborate with other partners to help it unlock new use cases for its image generator, including Canva, GoDaddy and Instacart.