OpenAI continues to lead the AI market thanks to the success of GPT-4. However, it seems that this is about to change. Google has just released Gemini, a new language model that vastly outperforms GPT-4 and all currently available AI technologies. Next, we’ll tell you why this is the most powerful AI and how to try it out.
As Google explains His blogGemini is a new multimodal language model that is the most powerful artificial intelligence to date. If you want to check it out, we encourage you to watch the video below where the Google team shows off everything it can do.
This AI has the ability to understand and generate text, visual, audio and code information. Hence its “multimedia” nature. This gives it superior properties compared to most existing language models (LLM). In fact, it scored the best in 30 out of 32 academic tests that test artificial intelligence, beating OpenAI’s GPT-4.
It even surpasses human intelligence, as Google says it scored 90% on the MMLU test, which covers up to 57 subjects including math, physics, history, law, medicine and ethics to assess knowledge and problem-solving skills. . .
What makes it different? Well, Gemini excels in their interactions without the need for a script. This makes it a more flexible experience for the user, who can communicate with the AI by talking to it or via video communication.
How to use Gemini:
Google releases Gemini in three versions: Nano, Pro and Ultra. The most powerful version is the Gemini Ultra and it will arrive in early 2024. The Nano is less powerful, but faster and is intended for simple tasks performed from the device.
Meanwhile, Gemini Pro is now available. Bard has updated and adopted this multimodal language model, although it is currently English only. So all you have to do is try Gemini to use Bard as built-in. Here we leave you – link: Gemini
– Read also: Meta launches new website, a free image generator with artificial intelligence. Be the first to use it