Google unveiled Gemini, a new AI model that is said to outperform Bard, ChatGPT, and other models

Google, a major player in the IT industry, has announced the release of Gemini, which it calls its largest and most powerful AI model.

The company says Gemini can function effectively on everything from data centers to mobile devices in a statement provided to wakenews. According to Google, the new AI model’s cutting-edge features will greatly improve how developers and enterprise clients use AI to build and scale their applications.

Sundar Pichai, the CEO of Alphabet and Google, believes that every technological advancement presents a chance to further human progress, promote scientific understanding, and enhance lives.

Additionally, according to Sundar, AI can open doors that could spur new waves of innovation, economic advancement, and knowledge-gathering. For these reasons, he and his group are eager to make AI useful to everyone on earth.

What’s relevant to Google Gemini

Gemini is described by Google as:

“Our most advanced and versatile model to date, exhibiting cutting-edge results across numerous industry benchmarks.” The Ultra, Pro, and Nano variants are all optimized for the original version, Gemini 1.0.

Pro is the model for scaling across a broad range of jobs, Nano is an effective model for on-device tasks, and Ultra is the largest and most capable for extremely complicated tasks.

Because it was designed to be multimodal, the new AI model can comprehend, operate on, and combine many forms of information, such as text, code, audio, images, and videos, while also being able to generalize. Furthermore, the model is said to be Google’s most adaptable model, meaning it can operate effectively on a variety of platforms, including mobile devices and data centers.

The AI program is made to effortlessly comprehend and process a wide range of inputs. Additionally, Gemini 1.0’s advanced multimodal reasoning abilities can aid in deciphering intricate written and visual data. Because of this, it is exceptionally adept at locating information that can be challenging to find among massive volumes of data.

Furthermore, the 1.0 version was trained to simultaneously recognize and comprehend text, photos, audio, and other media; as a result, it is more adept at understanding subtleties in information and can respond to inquiries on complex subjects. Because of this, it excels at clarifying thinking in challenging disciplines like physics and maths.

In the end, this initial version can comprehend, elucidate, and produce top-notch code in the most widely used programming languages worldwide, including Python, Java, C++, and Go. It is one of the most popular foundation models for coding in the world because of its capacity to function across languages and reason about complex data.

Accessibility

According to the statement, several Google products and platforms are currently implementing Gemini 1.0. Bard will start using an improved Gemini Pro version today for more sophisticated comprehension, planning, reasoning, and other functions.

More than 170 countries and territories will have access to Gemini in English, and Google intends to add support for additional languages and modalities soon.

In addition, Google is integrating the new AI model into Pixel, and in the upcoming months, it will be accessible in additional Google services and products, including Chrome, Ads, Search, and Duet AI. The internet giant claims to have already begun experimenting with Gemini in Search, where it is reducing latency in English by 40% in the US while simultaneously improving quality and speeding up the Search Generative Experience (SGE) for users.

In the end, Gemini Pro will be available to developers and enterprise clients using the Gemini API in Google Cloud Vertex AI or Google AI Studio as of December 13. Additionally, Android developers will be able to utilize AICore, a new system feature coming with Android 14, to build with Gemini Nano, Gemini’s model for on-device tasks, beginning with Pixel 8 Pro devices.

Before releasing Gemini Ultra to the general public, Google is presently finishing up safety and trust assessments and improving the model.

Before releasing Gemini Ultra to developers and enterprise customers in the first part of next year, it will first be made accessible for early experience and input to a restricted group of customers, developers, partners, and safety and responsibility specialists.

Author

Samson Ayodeji

I am literally obsessed with technology. I love writing about the latest news in technology in different markets, segments and sectors across the world.

View all posts