Top News

New Delhi: OpenAI has introduced its latest version of its AI image generation model and netizens have gone all crazy. Announcing the latest version of ChatGPT Images 2.0, the company said: “ChatGPT Images 2.0 is the next step: a state-of-the-art model that can take on complex visual tasks and produce precise, immediately usable visuals”. 

OpenAI stated that this new model is able to process instructions and incorporate detailed elements in images. It can understand small texts, icons and user interface elements far more accurately than its earlier version. This has been a challenge for its earlier image models.

“This model is a step change in detailed instruction following, placing and relating objects accurately, and rendering dense text, with the ability to generate across aspect ratios. Its sense of composition and visual taste means results feel less AI-generated and more intentionally designed. It’s accurate across languages and uses its expanded visual and world knowledge to fill in the gaps for you, so you get smarter images with less prompting,” OpenAI announced.

While netizens have widely appreciated this new model, some of them said that it is still “very raw” for geography.

Eugene Voroniuk, an X user wrote, “ChatGPT Images 2.0. are better than before. However, still very raw for geography/maps etc.”

One of the users was thrilled at the AI model's capability to generate manga and ads.

An X user, Otchy wrote, "Saw the title and thought, "Ugh, more exaggeration lol…" But looking at the examples, this isn't that much of an exaggeration, is it? / "So amazing it's hilarious, what is this—OpenAI officially announces powerful image generation AI for manga and ads, 'ChatGPT Images 2.0'"

Another wrote that while ChatGPT Image-2 was consistencies across “multiple generations,” the details have a “lot of artifacts”.

“ChatGPT Image-2 was pretty good in terms of the subject knowledge and with great consistencies across multiple generations. But currently the details seems to have a lot of artifacts,” an username who goes by the username idkpangea said on X.

An X user who used the AI model to generate a banner, said, “Just one prompt in ChatGPT → full moodboard + brand kit for a fictional girls clothing brand. Color palette, typography, packaging, retail, e commerce, social… all in one image. This is wildly good for a first round.”

ChatGPT Images 2.0 uses a wider range of styles than its earlier model. These includes photorealistic images, illustrations cinematic visuals, thereby letting users generate images in several aspect ratios. These make the genrated images apt for social media posts, presentations and banners. Users have expressed that the AI model mostly comes up to their expectations.