Shulou(Shulou.com)11/24 Report--

CTOnews.com, May 21, according to the official news of Asustek, Taiwan Zhiyun, a subsidiary of Asustek, recently released the enterprise-level large-scale language model "Formosa Model" at the annual meeting of AIHPCcon AI supercomputing.

▲ source: Shi Chongtang, chairman of Asustek's website, said that since the emergence of generative AI last year, Asustek has observed that corporate customers want to use generative AI to carry out a new wave of productivity revolution. CTOnews.com learned in the Asustek blog that Asustek's AI engineering team has been working with Nvidia since 2018 to participate in the development of Suan II, III and IV supercomputers, and invested in the establishment of Taiwan Zhiyun in 2021, becoming the only cloud service provider in Asia to provide commercial AI supercomputing services.

Besides releasing the Formosa model of OpenAI, Taiwan Zhiyun also combines AIHPC computing power and No-code large model optimization platform to launch a comprehensive enterprise-class generative AI solution, which provides a series of services such as fast transfer of AI supercomputer computing power, easy optimization model and elastic deployment model. It includes large-scale language model optimization service, large-scale language model optimization sharing service, global first large-scale language model local deployment solution, large-scale language model cloud hosting service and so on.

Shi Chongtang said that the Formosa model released by Zhiyun, a subsidiary of Asustek, was built with 176 billion parameters, matching the GPT-3.5 model of OpenAI, the parent company of ChatGPT. Some people in the industry think that the model trained in Chinese may be more suitable for local enterprises.

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.