Google has officially launched Gemini, heralded as its most robust generative artificial intelligence model to date. Boasting advanced multimodal reasoning capabilities, Gemini is primed for diverse applications across data centers to mobile devices, offering three distinct use cases—Ultra, Pro, and Nano.
Gemini’s rollout is orchestrated through Bard’s AI assistant in a phased approach. From December 6, a customized Gemini Pro version will enhance Bard’s English capabilities, showcasing superior reasoning, planning, and understanding. Google plans an early 2024 release of an Advanced Bard version, unleashing Gemini Ultra with even more advanced features.
Prior to public release, Google rigorously tested Gemini Pro in industry-standard assessments, surpassing GPT-3.5 in 6 out of 8 benchmark tests. Gemini Pro within Bard exhibits heightened proficiency in comprehension, generalization, reasoning, coding, and planning, exemplified by creating an intricately detailed paper airplane.
Bard’s integration with Gemini Pro, primarily for textual cues, is now accessible to users, with support for additional modalities on the horizon. Initially available in English across 170+ countries, Gemini will expand to other languages in various regions, notably Europe, following its initial rollout.
The pinnacle of this AI innovation, Gemini Ultra, will grace an extended Bard version in early 2024. Tailored for ultra-complex tasks, Gemini Ultra swiftly interprets and responds to diverse information types—text, image, audio, video, and code. Security checks for Gemini Ultra in Bard Advanced are in progress, with a testing program involving trusted users set to commence soon, heralding widespread availability in the coming year.
