Google presents Gemini 1.0, a new multimodal artificial intelligence model capable of performing highly complex tasks.
Artificial intelligence is positioned as one of the biggest revolutions of the last decade, which is why the giants of the technology industry have rushed to bet on AI. This is the case of Google, which after launching Bard is starting to deploy its generative AI model, Gemini 1.0, with the aim of getting ahead of its rivals.
Google’s project with Gemini is to present a generative AI model that has been trained to behave as a human would. The launch has three sizes that will be presented in phases, first with “Nano” and “Pro”, which will be immediately integrated into the Pixel 8 Pro smartphone and the Google Bard chatbot respectively, and “Ultra”, the most advanced model of the three that is still in internal testing phases, and will not see the light of day until 2024.
As Google points out through a statement, this AI has been built with the goal of being multimodal, meaning that it can “generalize and understand, operate and combine different types of information, including text, code, audio, image and video.”
We believe in making AI helpful for everyone. That’s why we’re launching Gemini, our most capable model that’s inspired by the way people understand and interact with the world. #GeminiAI pic.twitter.com/gNG9ha9xMO
— Google (@Google) December 6, 2023
Google will integrate Gemini into its products and services.
Gemini Nano is the variant focused on efficiency for processing tasks directly from Pixel 8 devices, in order to boost its services to users. This will boost new features such as ‘Summarize’ in the Recorder app and the implementation of ‘Smart Reply’ in Gboard.
For its part, Gemini Pro will be integrated to Bard to update the company’s chatbot, as pointed out from Google “this is the biggest update to Bard since its launch. It will be available in English in more than 170 countries and territories, and we plan to expand to different modalities and support new languages and locations in the near future”.
In addition, starting December 13, developers and enterprise customers will be able to access Gemini Pro through the Gemini API in Google AI Studio or Google Cloud Vertex AI .
Finally, Gemini Ultra is crowned as the most powerful version of AI developed by the company to perform complex tasks, and in order to compete head-to-head with the latest version of OpenAI’s ChatGPT and is expected to be available from next year, when its model will be used to launch “Bard Advanced”, an enhanced version of the conversational chatbot.
The artificial intelligence race is still on
The company points out that the results of Gemini in its Ultra version are the best ever recorded by an artificial intelligence in 30 academic tests out of 32, thus surpassing OpenAI’s GPT-4 results.
In addition, with a score of 90.0%, Gemini Ultra is the first model to outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects such as mathematics, physics, history, law, medicine and ethics to test both knowledge of the world and problem-solving skills.
As Sundar Pichai, CEO of Google and Alphabet, points out through a statement, “The transition we are seeing now with AI will be the most profound of our lifetimes, far greater than the step before mobile devices or the Web. AI has the potential to create opportunities (from the everyday to the extraordinary) for people around the world.” In this sense Pichai emphasizes that artificial intelligence presents new waves of innovation and economic progress, in addition, “it will drive knowledge, learning, creativity and productivity on a scale never seen before”.