Imagen 2

Google Bard Adds Imagen 2 Image Generation and Takes Gemini Pro Model Global

Google Bard has added a new image creation tool powered by Google’s Imagen 2 text-to-image model in an update that also brings the Gemini Pro large language model to Bard globally in more than 40 languages. Gemini Pro, initially launched in English last December, enhances Bard’s abilities in understanding, reasoning, summarizing, and coding. Imagen 2, meanwhile, allows users to create photorealistic images from a typed prompt, with safety measures like SynthID watermarks to differentiate AI-generated images from human artwork.

Imagen 2 Gemini Pro

Users can simply describe an image to Bard in English and it will generate custom, photorealistic results in most countries. Imagen 2 understands both simple and complex prompts to bring ideas to life. Google Cloud developers can also now leverage Imagen 2 for enterprise applications through Vertex AI. The model’s integration enables businesses to create product catalogs, advertisements, and other images from text descriptions.

The Imagen 2 release isn’t exclusive to Bard. The model is also available across Search and Vertex AI, broadening its application for users, creators, and enterprises. Google is also testing a new text-to-image experiment. The ImageFX pilot turns text prompts into images, then enables and allows for creative image generation with simple text prompts, and then offers  expressive chips for easily modifying results. ImageFX is available in the AI Test Kitchen along with other AI experiments.

“Imagen 2 is powered by Google DeepMind’s latest text-to-image advancements via a diffusion-based model. This update delivers our highest-quality images yet, as well as improvements in areas that text-to-image systems often struggle with, such as rendering realistic hands and human faces and keeping images free of distracting visual artifacts,” Google DeepMind vice president Eli Collins explained in a blog post. “Imagen 2 has been trained on higher-quality, image-description pairings and generates more detailed images that are better aligned with the semantics of people’s language prompts. It’s more accurate than our previous system at processing details, and it’s more capable at capturing nuance — delivering more photorealistic images across a range of styles and use cases.”

The global expansion of Gemini Pro in Bard makes the chatbot a lot more powerful on a global scale. In addition, The Gemini Pro expansion includes Bard’s double-check feature. Double-check enables users to verify the information provided by Bard against web content and improve the reliability and accuracy of Bard’s responses. Users can click an icon, and Bard will evaluate if there is web content to back up its answers.

Bard has never been as widely adopted as tools like ChatGPT, but that may change soon. There’s evidence that Bard may soon be seen as a real rival for the most popular generative AI chatbot. The Gemini Pro integration has rocketed Bardto to second place on the Chabots Arena leaderboard. The Large Model Systems Organization (LMSYS) chart places only OpenAI’s GPT-4 Turbo model ahead, with Elo scores of 1215 to 1239, respectively.


Google Releases Gemini Generative AI Model Designed to Match and Exceed OpenAI’s GPT-4 and ChatGPT

Google Search Unveils Generative AI Image Creation Feature

Google Chrome Embeds Generative AI to Write Text, Organize Tabs, and Design Browser Experience