On June 18th, Google announced on its official website blog that it will launch a Gemini mobile application in India that supports English and nine Indian languages, including Hindi, Bengali, etc. In addition, it will also launch an Indian version of Gemini Advanced.
Google did not explain much about launching applications in India, but mentioned India's "strong mobile first culture" in its blog.
Gemini is a large model launched by DeepMind Labs, a subsidiary of Google. It was initially released as version 1.0 in December 2023, and later upgraded to Gemini 1.5 in February 2024, with stronger performance and a long context window of 1 million tokens. At the developer conference in May, Google made some upgrades to Gemini 1.5.
Google has announced that it will introduce the powerful features of the state-of-the-art large model Gemini 1.5 Pro into the Indian version of Gemini Advanced, which has a context window of 1 million tokens and is currently the longest among all large model chatbots. It can process and understand large amounts of information, from lengthy documents (up to 1500 pages), emails, to hours of videos and a large code base.
Gemini applications can have both text and voice conversations, and can also add images for assistance. For specific application scenarios, Google mentioned using specific ingredients to create recipes, creating social media titles, and taking photos of flat tires for instructions on tire replacement.