Google recently introduced its latest AI model, Gemini, which is available in three versions: Nano, Pro, and Ultra. According to internal sources, Gemini Ultra, the most powerful variant, surpasses OpenAI’s GPT in nearly all industry-standard benchmarks.
Insiders from the company have disclosed that Gemini Ultra excels in 30 out of the 32 widely recognized academic benchmarks designed for evaluating Large Language Models (LLMs). This achievement could potentially mark a turnaround for Google, as the tech giant has faced criticism for trailing behind its AI rival, OpenAI, whose GPT model is renowned as the industry’s most potent and widely adopted. Google has integrated its Gemini Nano and Pro AI models into existing products, with the upcoming release of Gemini Ultra expected in early 2024.
The AI capabilities of Gemini Nano are utilized in Google’s Pixel 8 Pro smartphone, while Gemini Pro is accessible through the Bard chatbot. Additionally, Gemini Pro will be made available to enterprise customers and AI studio developers through Google’s Vertex AI program. Eli Collins, Vice President of Product at DeepMind, the division behind the AI platform, notes that Gemini Ultra is equipped to comprehend “nuanced” information in diverse formats, including text, images, audio, and code. Collins disclosed that the application was developed using data obtained from publicly accessible web content. However, the company did not openly specify the exact sources utilized to train the AI.
- Misleading Video Alters Prime Minister Shehbaz’s Speech Targeting Political Rival, Not His Party - 21/04/2024
- ATC Grants Three-Day Remand for Suspects in Ichhra Bazaar Harassment Case - 19/03/2024
- Pakistan, India, and Bangladesh Lead World in Severe Smog Crisis, Exceeding WHO Guidelines - 19/03/2024
