Google has announced Gemini 1.0. It is a new generation of AI model that can handle multiple types of data and tasks, from text and images to code and audio. Gemini 1.0 is the product of a large-scale collaboration between Google’s DeepMind and Research teams. It is inspired by how humans understand and interact with the world, according to Google CEO Sundar Pichai.
“I’ve always believed that if we could build smarter machines, we could harness them to benefit humanity in incredible ways,” Pichai wrote in a blog post.
Gemini 1.0 is designed to be natively multimodal. In simpler terms, it means it can process and combine different kinds of information without relying on separate components or systems. This gives it an edge over existing multimodal models as it can perform better on complex reasoning tasks, such as physics problems.
Gemini 1.0 is also proficient in coding and can generate and understand code in popular programming languages like Python, Java, C++ and Go. Google has used a specialized version of Gemini 1.0 to create AlphaCode 2. It is an improved version of its AI code generation system that can solve competitive programming challenges.
Gemini 1.0 comes in three different sizes: Nano, Pro and Ultra. Nano is the smallest and most efficient and is meant for on-device tasks. Pro is more versatile and scalable and will be integrated into many of Google’s products and services, such as Bard, Search, Ads, Chrome and Duet AI. Ultra is the largest and most capable and will be available for select customers, developers and partners next year.
Google has launched an enhanced version of Bard, which is fine-tuned with Pro. This upgraded version is expected to provide advanced reasoning, planning, and understanding capabilities. The enhanced Bard chatbot is accessible in the same 170 countries and territories where the regular Bard is currently available. Google also has plans to increase the availability of this new version throughout 2024.
In 2024, with the introduction of Gemini Ultra, Google will launch Bard Advanced, a more powerful AI with additional features. The capabilities of Pro will also be available through API calls via Google AI Studio or Google Cloud Vertex AI. In the upcoming months, Gemini functionality will be integrated into Search (specifically SGE), Ads, Chrome, and Duet AI.
Google claims that Gemini 1.0 is the most capable and general model it has ever built and that it surpasses the state-of-the-art performance in nearly every domain. Google also says that it is committed to developing and deploying Gemini 1.0 responsibly and safely and that it has conducted extensive safety evaluations and testing for potential risks and issues.Get latest Tech and Auto news from Techlusive on our WhatsApp Channel, Facebook, X (Twitter), Instagram and YouTube.
Author Name | Om Gupta