GOOGLE’s track record for creating and introducing technology comes with a lengthy and ever-growing list of specifications and capabilities that can leave some users feeling dazzled and even a little dizzy when it comes to trying to understand and interpret them all, even as they prove to be popular and hugely successful. This article aims to demystify the technical (and non-technical) concept of Gemini AI, making sure you’re empowered and informed so you can keep up with GOOGLE's latest AI invention.
At its essence, Gemini refers to a duality – a chatbot that is capable of performing a wide array of tasks, paired with a family of large language models (LLM) that underlie these capabilities. For readers overwhelmed with techno-gibberish, think of Gemini as GOOGLE’s homegrown version of ChatGPT, with a twist.
Diversity makes Gemini appealing: the device comes in three separate models, the Nano, Pro, and Ultra, weighing in at 180g, 350g and 900g. The Nano is the most cost-effective, lightweight model, suited to everyday tasks. On the other end of the spectrum, the Ultra variant is for the most demanding applications of artificial intelligence.
Even more confusing is how GOOGLE is trying to embed Gemini into all of its product lines. When it says that Gemini will be integrated into devices such as the Pixel 8, it’s referring to the AI model being used to improve features such as Smart Reply options, not the ability to just chat with it.
Gemini Nano is the GOOGLE AI that sits at the edge of its AI ecosystem. Designed to run efficiently on-device, such as for localised use in features such as Magic Compose.
While the front-end experience of Gemini Pro is essentially the same as the standard chatbot, the intermediary model opens up many more possibilities – and significant investment by GOOGLE has led it to claim parity with, and in many respects superiority to, OpenAI’s GPT-3.5.
The biggest of those, called Ultra, is the peak of what Gemini sells, but also stands for the pinnacle of GOOGLE’s AI as compute-intensive, and so far free, features are paywalled behind the $20 monthly subscription price.
GOOGLE births Gemini 1.5 Pro, the latest in a lightning-quick AI lineage. Like its parent, GOOGLE is doubling down on its competitive edge.
Beyond compute power, the Gemini Advanced subscription moves into a white-glove service, which includes access to the best that GOOGLE’s AI system has to offer.
While the company attempted to inject productivity-boosting AI into its GOOGLE Workspace platforms, it has now launched Gemini for Workspace.
As a part of this broader shift toward AI-powered art, Gemini has its own image generation tool.
Forty languages are supported by Gemini’s chatbot, but full feature availability differs from country to country.
The international scope of Gemini gives way to national borders when it comes to regulatory restrictions.
So, in the end, what can we say about GOOGLE’s Gemini project? To me, it’s a kind of techno-philosophical Magnum Opus, a blend of Wild Ambition with sheer technical brilliance.
© 2024 UC Technology Inc . All Rights Reserved.