Gemini launches personalized chatbots and advanced rendering model

Google launched two new features this Wednesday (28). GeminiThe company’s artificial intelligence (AI) platform is now more personalized and has gained new capabilities for image creation.

The first innovation, Launch of “Gems”, personalized chatbotsAdvanced, Business, and Enterprise plan users can now create their own AI versions, specializing in certain topics or a certain personality.

Jewellery They may specialize in tasks such as reviewing code, helping inspire planning, learning languages, or summarizing content.. It can also act as a companion in games like Q&A. The process is similar to custom GPTs already available in an OpenAI store.

To use Gems, simply click on the create chatbot button and write instructions for the Gem. Then just give it a name and start chatting. Moreover, Google has released some pre-configured examples for those who lack creativity or want to test the source first.

Although the tool has been released in Brazil, it is currently only possible to create Gems in English.

New Picture 3

The second member of Gemini is Launch of the company’s most advanced imaging model, Imagen 3It will now be available to all users regardless of their Google account type.

Now users will be able to get more accurate AI image results based on specific text prompts. For example, they can even include the style of the illustration, such as photorealistic or oil paintings.

Google ensures that its language model has built-in protections and uses SynthID to apply invisible watermarks to creations and “tell” them that they are synthetic.

In the coming days, Gemini Advanced, Business, and Enterprise account users will continue to have access to: creating images of people in advanceThis is still an experimental feature and is only available in English, with rules in place to ensure the images are not used in a criminal manner or to infringe copyright.

“We do not permit the creation of photorealistic and identifiable images of individuals.“representation of minors or excessively bloody, violent or sexual scenes,” Google says. The features will be rolled out gradually and will be available to the community very soon.