Last night, Google finally iroduced the Gemini artificial ielligence model, which is considered the company's biggest and best artificial ielligence model. This model has been designed as multimodal from the beginning so that it can understand and analyze various data such as text, image, video and sound. On the other hand, Google has paid special atteion to providing detailed explanations about complex math and physics issues.
This model is available in three sizes: Nano, Pro and Ultra. Ultra is the largest size of this model, which is used for highly complex processing. Pro is the right version for many tasks, and general users can take advaage of this version's features, and finally, Nano is designed to run the model using the phone's hardware.
Naturally, the most attractive version is the Ultra version, which has managed to perform better than GPT-A in 30 of the 32 importa benchmarks of artificial ielligence models. This model works better than OpenAI's GPT-4 in terms of programming, so users can get better codes by using it.
Google has announced that the Bard chatbot is now using Gemini Pro, so users can start using it. Of course, Gemini Pro is currely only activated for English language in Bard, but soon other languages will also receive this new feature.
On the other hand, Google has announced that Gemini Ultra will soon be preseed in a new version of Bard called Bard Advanced, which will provide users with more advanced features. Of course, Bard Advanced probably won't be offered for free. Regarding Gemini Nano, we must say that this feature is currely only available for the Pixel 8 Pro, and it is not clear whether other new Pixel phones will receive this feature in the future.
Google has announced that in the coming mohs, the Gemini artificial ielligence model will be widely implemeed in various services of this company to significaly improve the user experience.
Source: GSMArena
