Home technology artificial-intelligence Imagen 3 AI Model for Image Generation is added to Google's Gemini
Artificial Intelligence
CIO Bulletin
2024-10-11
All users and developers can now use Google’s Gemini AI model with upgraded image generation capabilities from Imagen 3, Google has announced.
On Wednesday, everyone saw the introduction of an update, which intends to improve the visual features of Gemini. It will enable free-tier users and developers to utilize Imagen 3 through the Gemini API. The model in question is intended to enhance the photorealistic aspect of the model, follow user instructions closely and reduce the presence of unwanted objects in the generated images.
In a recent update shared via the social media platform X (also known as Twitter), Google displayed the performance of Imagen 3 as one of the advancement which is better than its rivals like Meta AI. In example, Gemini has superior resolution of 2048 by 2048, higher than Meta’s 1280 by 1280, in response to a command requiring an image of a golden retriever dog seated inside of a moving train.
Imagen 3 features several ways for you to express yourself; this ranges from photorealism to textural oil painting and Claymation styles. Users can also indicate camera styles like ‘Nikon DSLR or GoPro’ thus enhancing the creativity.
To allay fears that AI-generated content may be used improperly, Google has rolled out SynthID technology that adds a covert watermark to every image. This feature aims to stop modifications as well as viral deep fakes.
As Imagen 3 is now available, every person and developer will be able to more than ever enjoy Google’s world-class features in AI image generating, thus increasing the creative scope entailed in Gemini app and API.
Digital-marketing
Artificial-intelligence
Lifestyle-and-fashion
Food-and-beverage