Google has announced the discontinuation of Bard, roughly a year after its initial launch aimed at competing with OpenAI’s ChatGPT. The service is undergoing a transformation, adopting the new name Gemini, which aligns with the company’s foundational model series. In a significant development, Google is introducing Gemini Ultra, its most advanced large language model to date.
However, accessing Gemini Ultra will come with a cost. Google plans to offer it as part of a newly introduced $20 Google One subscription tier. This package not only includes a two-month trial period but also provides subscribers with 2TB of storage and full access to Google One’s range of features.
Additionally, subscribers will gain entry to Gemini within Google Workspace applications such as Docs, Slides, Sheets, and Meet. Concurrently, Google will phase out the Duet AI brand, which was previously limited to AI functionalities within Workspace, integrating these features into Gemini as well.
Additionally, Google is unveiling a Gemini application for Android users and integrating it into the Google application on iOS. Android users will have the option to substitute Google Assistant with Gemini as well. Google has also indicated plans to offer API access to the Ultra model for developers, promising more details in the near future.
The Gemini Advanced service will roll out across over 150 countries and territories, initially supporting only English. However, Google plans to expand its linguistic capabilities, with Japanese and Korean slated as the next languages to be added to the platform.
Here is how Google’s Sissie Hsiao explains the new Ultra model: “With access to our Ultra 1.0 model, Gemini Advanced is far more capable at highly complex tasks with a range of applications like coding, logical reasoning, following nuanced instructions and creative collaboration, amongst many others. Not only does Gemini Advanced allow you to write longer prompts, it can also better understand the context of your previous prompts.”
It is also worth mentioning that Gemini Ultra is a multimodal model, meaning it can also hold conversations based on images and other means of input.
