New Delhi, April 22 (IANS). OpenAI, the leading artificial intelligence (AI) and parent company of ChatGPT, has launched ‘Images 2.0’ in its platform. It is a next generation image generation model, which aims to provide accurate, real and more usable visuals for greater reasoning capabilities.
The company said that Images 2.0 improves the way it follows instructions and handles fine elements in images. It renders small text, icons and user interface elements more accurately than before.
It is also capable of generating images with different proportions, which include social media graphics and presentations.
According to the company, it has the potential to be a major update. The model is able to use web search models for real-time information and can generate multiple different images from a single prompt.
According to OpenAI, ChatGPT Images 2.0 is also capable of performing in different languages. This model can reproduce text inside photos in Japanese, Korean, Chinese, Hindi and Bengali languages with greater accuracy.
In terms of visual quality, Images 2.0 delivers truer and more accurate visuals in formats such as photographs, cinematic stills, manga and pixel art, as well as better handling of lighting, textures and fine details.
The company said that the Images 2.0 UI simplifies design workflows across a variety of uses and styles such as screenshots, magazine layouts, infographics, hand-written notes, comics and manga, advertising and cinematic visuals, as well as on platforms such as Canva, Figma and Adobe.
OpenAI said that this new model will be available on platforms like ChatGPT and Codex. However, some advanced features to think about will be available only to paid users.
Additionally, the company also revealed that the ‘GPT-Image-2’ model can be accessed through an API, the price of which will vary depending on the quality and resolution of the selected images.
–IANS
abs/
