Google unveils Gemini, its ‘most capable’ AI model to date Google unveils Gemini, its ‘most capable’ AI model to date
Gemini Ultra marks Google's leap forward in multimodal AI technology, setting new benchmarks in AI language and coding proficiency.
2 min read
Updated: Dec. 6, 2023 at 8:57 pm UTC
Cover art/illustration via CryptoSlate. Image includes combined content which may include AI-generated content.
On Wednesday, Google unveiled Gemini, its most advanced artificial intelligence system to date, marking a significant milestone in its AI development.
Dubbed “our largest and most capable AI model yet” by Google CEO Sundar Pichai, Gemini is designed to understand and operate seamlessly across text, images, audio, video, and other modes of information.
“This new era of models represents one of the biggest science and engineering efforts we’ve undertaken as a company,” wrote Pichai in the announcement. “I’m genuinely excited for what’s ahead, and for the opportunities Gemini will unlock for people everywhere.”
The company said the new system, optimized in three sizes, greatly enhances Google’s AI offerings and outperforms previous benchmarks. Gemini scored higher than any other model on over 30 leading AI benchmarks, according to Google’s tests.
Most notably, Gemini Ultra, the largest version, became the first AI system to surpass average human performance on a language test spanning over 50 academic subjects. The statement continued;
“With a score of 90.0%, Gemini Ultra is the first model to outperform human experts on MMLU (massive multitask language understanding), which uses a combination of 57 subjects such as math, physics, history, law, medicine and ethics for testing both world knowledge and problem-solving abilities.”
Gemini also displayed new abilities in computer programming. Using a specialized version of Gemini, Google created an advanced code generation system called AlphaCode 2, which it says “excels at solving competitive programming problems.”
When evaluated in programming competitions, AlphaCode 2 solved nearly twice as many problems as its predecessor, with Google estimating it would outperform 85% of human participants.
Google is immediately releasing Gemini on several of its products, including its Bard conversational AI tool launched earlier this year and Pixel smartphones.
“Starting today, Bard will use a fine-tuned version of Gemini Pro for more advanced reasoning, planning, understanding, and more. This is the biggest upgrade to Bard since it launched,” said Google.
Pixel 8 Pro also became the first phone enabled with Gemini Nano, powering new on-device features like text summarization in apps.
Select developers and partners will gain early access to Gemini Ultra later this year before it becomes widely available through Google’s developer platforms in 2023. Google said it is currently completing safety checks on the model.
Mentioned in this article
Author 
Jacob Oliver
Jacob Oliver is a recovering academic and English teacher turned crypto journalist and web3 writer. He holds a Ph.D. from the University of Washington.
Latest US Stories
Featured Story
Advertise Here
In this article
Google LLC is an American multinational technology company that is most commonly known for its search engine that was founded by Sergey Brin and Larry Page in 1998.
Bard by Google
Google’s Bard is a conversational artificial intelligence chatbot based on the LaMDA family of big language models.
