Google's rebranded AI platform Gemini includes new app and subscription service

The service formerly known as Bard is now available as an app on Android

Sundar Pichai, chief executive of Google and its parent company Alphabet. AP

Google has rebranded its generative artificial intelligence platform Bard – and it's more than just a name change.

The Alphabet subsidiary says the platform will now be known as Gemini, as it seeks to further challenge OpenAI's ChatGPT and grab a bigger slice of the global generative AI market.

What is Gemini?

For background, Gemini is also the name of Google's suite of large language models powering Bard that was introduced in December – at the time the platform's biggest update.

Coming with multimodal reasoning capabilities, the Gemini 1.0 model came in three sizes – nano, pro and ultra – so it can run on everything from resource-intensive data centres to small mobile devices.

This time, Gemini is bigger than ever – and Advanced.

Gemini Advanced is Google's highest-tier generative AI offering, featuring its largest model Gemini Ultra 1.0, which it describes as its "most capable" AI model.

It is "far more capable for reasoning, following instructions, coding and creative collaboration", it "can understand, explain and generate high-quality code in many programming languages" and is designed for highly complex tasks.

How powerful is Gemini Ultra 1.0?

Gemini Ultra 1.0 is the first to outperform human experts on massive multitask language understanding, "evolving to be more than just the models" and "supports an entire ecosystem", Sundar Pichai, chief executive of Google and Alphabet, said in a blog post.

Gemini Ultra 1.0's MMLU uses a combination of 57 subjects – including maths, physics, history, law, medicine and ethics – to test knowledge and problem-solving abilities, according to Google.

Gemini is also available in 40 languages but Google plans to add more, including Japanese and Korean, to this list.

How much will it cost?

For those willing to shell out cash for Google's most powerful AI model, it'll cost you $19.99 a month on the new Google One AI Premium plan – and you'll be charged only after a free two-month trial.

The plan includes 2TB of storage, more features for Google Meet and Google Calendar, and 10 per cent back from purchases made on the Google Store.

The subscription will also grant users access to future Gemini features that will be made available on Gmail, Google Docs, Google Slides and Google Meet.

How about the app?

Google said the Gemini app for Android is now available for download, while those using Apple's iPhones can use the service through the Google iOS app.

The company did not give a date for the iOS launch.

Is Google far off from OpenAI?

The rise of generative AI triggered a race among technology companies to bring the powerful technology to more enterprises and consumers.

Google's move to introduce an app and paid tier is expected to raise the stakes, most notably in its battle with Microsoft-backed OpenAI.

The world's biggest internet company, however, may still have some catching up to do, as OpenAI has built a lead on both fronts: San Francisco-based OpenAI began making the ChatGPT app available in May and a new premium service, ChatGPT Enterprise, in August.

Last month, OpenAI launched the online ChatGPT Store, which gives users access to more than three million custom versions of generative trained transformers to help them customise their chatbots and allow them to earn money.

Expect more from Google

Google is continuing to boost its generative AI platform, "already well under way training the next iteration of our Gemini models", said Mr Pichai.

Next week, he said, the company will share "more details" for the cloud and developers, noting that "hundreds of thousands of developers and businesses have already been building with Gemini models".

"Developers have been fundamental to every major technology shift and will play an equally important role in the Gemini ecosystem," Mr Pichai said.