The function to generate and edit images has been improved in GPT Image 1.5.
OpenAI has introduced GPT Image 1.5, a new AI model specifically optimized for generating and editing images, designed to be faster and more cost-effective than its predecessor.
Faster and more efficient
According to OpenAI, ChatGPT can generate images up to four times faster with the new model. This indicates a more efficient use of hardware, which not only increases speed but also reduces costs and energy consumption. In addition, the API prices are approximately 20 percent lower than with the previous image model.
Better control over details
A key improvement is the way the model handles edits. With previous AI image models, unwanted details often changed, such as logos or specific elements. OpenAI says that the latest model is better at remembering relevant details while implementing the requested adjustments, which is especially important for professional and commercial use.
The model can also perform more complex tasks, such as combining objects from multiple images and then displaying them in a different style.
Improved text display
Like recent competing models, GPT Image 1.5 places more emphasis on correct text display in images. The model is better at rendering small and compact text, making it more suitable for applications such as diagrams, presentations, and infographics.
GPT Image 1.5 is available via an API for developers and via the image function in ChatGPT for end users.