Network Security Internet Technology Development Database Servers Mobile Phone Android Software Apple Software Computer Software News IT Information

In addition to Weibo, there is also WeChat

Please pay attention

WeChat public account


Google high-profile bomb field: the official debut of the new big language model Gemini 1.0, almost completely ahead of OpenAI GPT-4

2024-07-21 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > IT Information >


Shulou( Report--

Thanks to netizens, soft media users 1520111, Baa Yang, Alejandro86 clue delivery!, December 6 (Xinhua)-- Google CEO Pichai revealed the existence of Gemini for the first time at the I / O developer conference in June and is now officially available to the public. Google says this is its most powerful and versatile large language model to date.

According to Google's benchmark results, Gemini showed "state-of-the-art performance" in many tests and even completely beat OpenAI's GPT-4 in most benchmarks.

Google has released three different versions for different scenarios: Gemini Ultra, Gemini Pro and Gemini Nano. These different versions will also apply to different scenarios and cases:

Gemini Ultra-the strongest and largest model for handling highly complex tasks.

Gemini Pro-the best model for extending various tasks.

Gemini Nano-the most efficient model for devices such as mobile phones.

Starting December 13, developers and enterprise customers will be able to access Gemini Pro's Gemini API through Google AI Studio or Google Cloud Vertex AI.

Gemini Ultra currently only provides early experiments and feedback to invited customers, developers, partners and security experts, and plans to launch it to developers and enterprise customers early next year.

Comparison between Gemini and GPT-4: Gemini Ultra's performance shows "advanced results" beyond the current era in 30 of the 32 widely used academic benchmarks, which are also the most commonly used and widely used tests in the field of LLM.

Among them, it became the first model to surpass human experts in MMLU (large-scale multitasking language understanding) with a score of 90.0% (the test combines 57 disciplines such as mathematics, physics, history, law, medicine and ethics).

Gemini Ultra also showed a 59.4% lead in the new MMMU benchmark, which covers multimodal tasks in different areas that "need to be thought out."

Google said its Bard would be enhanced more advanced with the Gemini Pro version, involving logical reasoning, planning and understanding functions. It will provide services in more than 170 countries and regions and plans to expand to different modes and support more languages and regions in the future.

In addition, a new upgrade of Bard Advanced will be launched early next year, so that users can better experience its best model and functionality, namely Gemini Ultra.

According to reports, Android developers will also be able to build Gemini Nano with AICore ( Note: a new system feature provided in Android 14), while terminal devices will be the first to support Pixel 8 Pro series models.

In the coming months, Gemini will be applied to more products and services, such as search, advertising, Chrome, Duet AI and so on.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.

Views: 0

*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.

Share To

IT Information


© 2024 SLNews company. All rights reserved.