The next artificial intelligence model from Google, Gemini, will be able to handle a wide variety of data kinds, including text, photos, and graphs, thanks to its multimodal capabilities.
Google revealed ‘Gemini,’ its next-generation AI model, at Google I/O 2023. Google was supposed to introduce its GPT-4 competitor next week, but a new rumor says January 2024. The Information reported that Google CEO Sundar Pichai delayed the release date to January next year due to non-English queries.
Google needs multilingual support to match OpenAI’s GPT-4, and the report says it has “met that standard in some respects.”
A User Shares The Video Link:
Google delays the rollout of Gemini, their ChatGPT competitor.
When?
It will launch in January.
Why?
Some defects in non-english languages.
Hate that there’s a delay but companies don’t want to launch if they know it won’t shake the internet.pic.twitter.com/WFoa67YtUD
— Aadit Sheth (@aaditsh) December 4, 2023
Last month, Pichai said the business is concentrating on releasing Gemini 1.0 quickly but is still finalizing the largest next-generation AI model.
Gemini: What is it?
Google made it plain during this year’s Google I/O that Gemini would be the most potent model the firm has worked on when it debuted it in May. Gemini, developed by Google’s Brain Team and DeepMind in collaboration with PaLM 2, will reportedly be “multimodal,” or capable of processing and storing a variety of media formats (including text and graphics), as well as planning and memory tasks. Google will use Gemini across all of its products, spread it out to developers and cloud customers, and mix many AI models. The next-gen AI model will be “highly efficient with tools and API integrations,” according to Pichai. Existing AI-powered Google products, such as Search, Google Assistant, and the famous chatbot Bard, are anticipated to be driven by the next AI model.
Causes of delay due to non-English inquiry
The reason behind the delay in Gemini’s release was allegedly because Google “found the AI didn’t reliably handle some non-English queries.” In November, Google CEO Sundar Pichai stated that the company’s focus is on releasing Gemini 1.0 quickly and ensuring it is competitive enough, emphasizing the significance of worldwide language compatibility. According to sources, Gemini has either caught up to or perhaps exceeded GPT-4 in certain respects despite the language barrier. Curiously, Gemini intends to surpass OpenAI’s model in terms of its sophisticated picture and text-generating skills.
The AI model’s multimodal capabilities are quite remarkable.
Previously unseen in earlier models, Gemini’s remarkable multimodal skills were unveiled by Google at I/O 2023. In addition to effectively integrating tools and APIs, the AI model can read text and graphics. Gemini, according to Google VP and Bard and Google Assistant manager Sissie Hsiao, could “draw me three pictures of the steps to how to ice a three-layer cake” when prompted. Meanwhile, rumor has it that Google is supposedly still “finalizing the primary, biggest version of Gemini.”
Google service integration and future developments
The integration of Gemini into Google products like Bard, Search, and Workspace is still in the works. Nevertheless, a lightweight “Gecko” version of Gemini will be available for mobile devices, among others. “Enable future innovations, like memory and planning.” is another way the business described Gemini. Techies are becoming excited about how Gemini’s advanced features will affect AI applications as the launch date draws near.