Google’s Gemini has introduced a series of exciting updates, including personalised AI features and improved image generation capabilities, powered by the latest Imagen 3 model.
Dave Citron, Senior Director of Product Management at Gemini, announced these enhancements in a recent statement. Google has also released a range of pre-configured Gems for different purposes, including Learning Coach, Brainstormer, Career Guide, Writing Editor, and Coding Partner.
Imagen 3 is touted for its high-quality image generation, capable of creating visuals from minimal textual input. Users can request various styles, such as realistic landscapes, textured oil paintings, or playful claymation scenes.
Imagen 3 includes advanced features and built-in safeguards that align with Google’s design principles. The model excels in performance compared to its predecessors. As with Imagen 2, the SynthID watermarking tool for AI-generated images will be used.
In the coming days, Google will start offering early access to Imagen 3’s capability to generate images of people for Gemini Advanced, Business, and Enterprise users, initially in English. Significant technical improvements and rigorous evaluations have reportedly been implemented to enhance this feature.