Google just released a new AI model called Gemini

← Prev

On December 6th, Google launced a new and more powerful AI model called "Gemini". It is available in Bard. These are some of its key features.

Google just released a new AI model called Gemini

a) Gemini is trained to understand our ways of life and might even behave like human.

b) It can take text inputs like most models. But it can also code as well as take inputs as audio, video and image.

c) According to Google, Gemini has the potential to beat ChatGPT and GPT-4, since Gemini tops 30 out of 32 performance benchmarks, which includes, 10 out of 12 popular text and reasoning benchmarks, 9 out of 9 image understanding, 6 out of 6 video understanding and 5 out of 5 speach recognition and translation benchmarks.

d) It will be available in 3 versions.

1. "Gemini Nano", a lighter version for smartphones like Android devices. 2. "Gemini Pro", is trained to do a broad range of tasks and its been already incorporated in "Bard", Googles AI powered chatbot. 3. "Gemini Ultra", will be more capable, for large models that will do complex tasks. It

According to Google, it is built with responsibility and safety at the core. Gemini has the most comprehensive safety evaluations of any Google AI model to date, including for bias and toxicity.

As of now, Gemini in available in English language only. Hopefully, it will come with other languages very soon.

I guess, its too early say that "Gemini" will dominate in the field of "AI" and will kill ChatGPT. But its definitely a game changer. The battle for "AI" supremacy has just begun.

We have entered into a new "AI" era. Let us hope, it will be good human kind.

You can read more about Gemini here.

← Previous