Bard has received Gemini Pro integration in all supported countries, as well as a free chat image generator based on Imagen 2

Mojahid Mottakin/Unsplash.

Google's Bard chatbot has received a major update - the company has integrated its latest language model, Gemini Pro, and added an Imagen 2-based image generation feature.

Here's What We Know

Previously, Gemini Pro in Bard was only available in English. Now, the chatbot can use the model in 40 languages in 230 countries and regions around the world. The full list can be found on Google's support site.

In addition, the double-check answers feature has worked in all languages too. If you click on the "G" icon under the result, Bard will check if there is any material on the Internet to support the answer.

Another major new feature is the inbuilt image generator based on Imagen 2, which is available in English in most countries.

When creating illustrations, Google recommends:

• Start the prompt with the words "draw", "generate", "create";
• Describe the desired style of image, for example: photorealism, charcoal drawing, watercolour painting, cartoon illustration.
• Include a detailed visual description of the desired image, for example: describe the object, its actions, and the background or environment.

For security, digital watermarks are embedded in the pixels. Thanks to this, Google image search will be able to recognise whether an image has been generated using AI. Also, the service will not create pictures with tooltips that contain people's names.

In parallel, Google announced ImageFX, a service designed for experimental generation of text-based graphics. For now, it is only available to English-speaking users in the US, Kenya, New Zealand and Australia.

Source: 9to5Google