OpenAI has released a new version of ChatGPT Images, introducing a flagship image generation model designed to produce images more quickly while offering finer control over edits, as the company steps up competition in generative artificial intelligence.
The update, powered by the new GPT Image 1.5 model, is being rolled out globally to all ChatGPT users and is also available through OpenAI’s application programming interface. OpenAI said the model can generate images up to four times faster than previous versions and is better at preserving details such as lighting, composition and the appearance of people across multiple edits.
See also: Google Unveils Disco as Experimental AI Browser Initiative
Alongside the model update, OpenAI introduced a redesigned Images feature within ChatGPT, giving users a dedicated space to create and experiment with images using preset styles and prompts. Business and Enterprise customers will receive access to the new Images experience at a later date, the company said.
“Now, whether you’re creating something from scratch or editing a photo, you’ll get the output you’re picturing,” OpenAI said in its press release, adding that the system makes precise edits while keeping important elements intact. The company said the model supports a wide range of edits, including adding, removing, blending and rearranging elements, as well as creative transformations that incorporate text and layout changes.
See also: OpenAI to Support BBVA’s Plan for AI-Driven Banking Operations
OpenAI said GPT Image 1.5 also improves instruction following, text rendering and overall image quality, making it more suitable for use cases such as marketing, branding and e-commerce. Image inputs and outputs are priced about 20% lower than the previous version, OpenAI said, allowing developers to generate and refine more images within the same budget.
The earlier version of ChatGPT Images will remain available as a custom option, OpenAI said, adding that the latest release represents an incremental step forward and that further improvements to image generation and editing are planned in future updates.
Source: OpenAI