Google upgrades Gemini with Imagen 3 AI model for image generation
Google has announced a significant update to its AI model Gemini, introducing its latest image generation model, Imagen 3, to enhance the visual capabilities of the Gemini chatbot. This recent upgrade, unveiled on Wednesday, is geared towards enhancing Gemini’s image-generation features, providing advanced tools to all users, including those utilizing the free tier. Developers utilizing the Gemini API will now have access to Imagen 3 for creating applications and experiences with improved visual capabilities.
Superior Image Generation with Imagen 3
Imagen 3, Google’s newest AI image generator, delivers superior photorealism, increased adherence to prompts, and reduced unintended elements in the generated images. According to Google's announcement on X (formerly known as Twitter), all Gemini app users can now leverage this advanced model to create images. Initial tests have demonstrated that Imagen 3 outperforms similar models like Meta AI. In a direct comparison, Gemini's Imagen 3 excelled in accurately rendering more details of an image of a golden retriever on a train, at a higher resolution of 2048 x 2048, surpassing Meta’s 1280 x 1280 resolution.
Artistic Styles and Camera Options
The Imagen 3 model supports various artistic styles, ranging from photorealism to textured oil paintings and even claymation. Users also have the option to specify a camera style for the image, such as a Nikon DSLR, a GoPro, or a wide-angle lens. This level of flexibility enables creative freedom across diverse visual aesthetics, catering to both casual users and professionals.
Enhanced Security Measures
With this update, security takes the spotlight. To address the misuse of AI-generated content, Google has incorporated SynthID technology, embedding an invisible watermark directly into the pixels of each image. This AI label is irremovable and resistant to cropping, even in screenshots, serving as a deterrent against the proliferation of deepfakes and unauthorized alterations.
Availability of Imagen 3 Model
The Imagen 3 model is now accessible within the Gemini app and API, simplifying the exploration and utilization of Google’s latest advancements in AI image generation for both developers and users.
For more information, you can visit Business Today Magazine
