Google has introduced Gemini, its latest large language model.
There will be a gradual integration of Gemini into Google’s ecosystem, including its search engine and Chrome browser. One version, Gemini Pro, is already powering Google’s Bard AI chatbot, and even Pixel 8 Pro users are getting advanced features enabled by another version rolling out called Gemini Nano.
CEO Sundar Pichai hinted at Gemini during theI/O developer conferenceearlier this year, and now Google is making it available to the public.
What is Assistant with Bard and what can it do?
The new version of Google’s famous Assistant is using Bard for extra help - find out what you should know, here.
More than a text processor
Yes, Gemini is Google’s response to OpenAI’s GPT models, but it’s also Google’s most advanced AI model to date, designed to be more than just a text processor. It supports multiple forms of data input, including text, images, video, and audio, and has the potential to be a significant leap forward from existing AI technologies. Google isn’t framing it as an incremental improvement in AI but rather a paradigm shift toward more integrated, multimodal AI systems.
“It supports multiple forms of data input, including text, images, video, and audio…”
While most GPT models primarily focus on text-based processing, Gemini is designed from the start to be natively multimodal, capable of understanding and processing multiple types of data inputs. This gives it a significant edge in tasks that involve audio and video interactions - something GPT-4 is only starting to do. In benchmark comparisons, Google claimed Gemini outperformed OpenAI’s GPT-4 in 30 out of 32 tests involving video and audio processing.
Staggered release dates
Gemini currently comes in three different versions, and its launch is staggered. The first version, Gemini Nano, is lightweight and runs natively and offline on Android devices. The second, Gemini Pro, is more robust and serves as the core for numerous Google AI services, including Bard, Google’s AI chatbot. The third, Gemini Ultra, is the most powerful version, designed primarily for data centers and enterprise applications.
Google said developers and enterprise customers will get access to Gemini Pro through Google Generative AI Studio or Vertex AI in Google Cloud on Dec. 13, and Gemini Ultra will debut next year. Although initially available in English, Google plans to extend its capabilities to other languages soon.
How to try Gemini Pro right now in Bard
Google’s Gemini AI is now available for you to try out for free within the Bard chatbot.
Just head over to the Bard website in your browser, log in, or create an account if you don’t already have one. Keep in mind that Bard specifically requires a Google account for access. If you’re using Google Workspace, you might need to switch to your personal email account to give Gemini a try.
This is still an experimental phase, so be prepared for the occasional software glitch in your chatbot responses. Also, remember that Bard is integrated with other Google services, so you can tag “@Gmail” to get message summaries or “@YouTube” for topics related to videos.
Want to try Gemini Nano on Pixel 8 Pro?
Pocket-lint has all the details on the latest Pixel Feature drop and new AI capabilities:
Google’s December Pixel Feature Drop brings new on-device AI features to the Pixel 8 Pro
In tandem with the launch of Google’s Gemini series of AI models, some Pixel owners can expect new tricks from their phones.
