Alphabet, the parent firm of Google, has launched its largest and most advanced artificial intelligence (AI) model to date called Google Gemini. The model is designed to rival competitors like OpenAI’s GPT-4 and Meta’s Llama 2. Gemini was initially teased by Google CEO Sundar Pichai at the Google I/O developer conference in June and is now being officially released to the public.
The Gemini AI model is available in three different sizes: Ultra, Pro, and Nano. The Ultra version is tailored for complex tasks, the Pro version is scalable across a wide range of tasks, and the Nano version is specialized for on-device tasks. One of the key features of Gemini is its multimodal approach, which enables it to process various forms of information simultaneously, including text, code, audio, image, and video.
Currently, Gemini is only available in English, but Google plans to expand its language support in the future. CEO Sundar Pichai envisions integrating the model into various Google products such as the search engine, ad products, and the Chrome browser. He believes that Gemini represents the future of Google and arrives at just the right time.
Gemini has been developed through a collaborative effort by teams across Google, including Google Research. Demis Hassabis, CEO and Co-Founder of Google DeepMind, explains that Gemini was built from the ground up to be multimodal, allowing it to seamlessly understand and combine different types of information. The AI model is expected to evolve beyond its current capabilities, incorporating additional senses and functionalities over time.
Google is rolling out Gemini through various channels. Bard, powered by Gemini Pro, will benefit from new features, while Google Pixel 8 Pro users will experience enhancements courtesy of Gemini Nano. The release of Gemini Ultra is scheduled for 2024. Starting from December 13, developers and enterprise customers can access Gemini Pro through Google Generative AI Studio or Vertex AI in Google Cloud.
Although Gemini’s basic models currently focus on text inputs and outputs, more advanced versions like Gemini Ultra will be capable of handling images, video, and audio. The availability of Gemini Ultra is currently limited to select customers, developers, partners, and safety and responsibility experts, who will provide feedback for further improvement. A wider release is expected for early next year.
Google’s goal is to continuously enhance Gemini, reducing any flaws such as biases or hallucinations, while improving its overall performance. As the AI model gains a deeper understanding of the world, its accuracy, awareness, and grounding will be further enhanced.
Google Gemini sets its sights on becoming a formidable competitor in the AI landscape, providing users with advanced capabilities and a multitude of applications. With its multimodal approach and future potential for growth, Gemini aims to solidify Google’s position in the rapidly evolving field of artificial intelligence.
(Note: This response has been generated by an AI language model. It has been reviewed and edited by a human editor to ensure accuracy, clarity, and adherence to guidelines.)
