Since the beginning of this year, domestic and foreign technology companies have been competing to expand their business landscape around large models. According to statistics, as of the end of July this year, there have been a total of 130 large models released in China, of which 64 were released in the first seven months of this year alone. Domestic large models have entered a period of rapid development, accelerating the digital and intelligent transformation of Qianxing Baiye. The Hangzhou Asian Games, which is currently underway, is the first ever cloud based Asian Games in history. Based on Alibaba Cloud, the core system is 100% cloud based. At the Digital Technology Experience Center of the Hangzhou Asian Games, tourists can use the art installation "Imagine the Future Asian Games" to create a large model of Alibaba Cloud's painting creation "Tongyi Wanxiang" for on-site painting, generating Asian Games themed images. Alibaba Cloud is one of the earliest technology companies to invest in the research and development of large models, and began laying out large models in 2019. Alibaba Cloud CTO Zhou Jingren told reporters that since the beginning of this year, Alibaba Cloud has continuously launched multiple self-developed Tongyi large models and large model application products, which have the ability to process or generate multiple modal content such as text, voice, and images, and are widely used in various industries. On September 21st, Huawei Cloud released the Pangu Medical Big Model at the Huawei All Connected Conference 2023. This model has the ability of medical clinical assistance, which will effectively support the application of scenarios such as medical testing, clinical assisted diagnosis and treatment, and personal health management. Huawei Executive Director and Huawei Cloud CEO Zhang Ping An introduced that Runda Medical Model, Liangyi Xiaohui, based on the Pangu model, will visit 4000 hospitals to assist doctors in interpreting and diagnosing reports. On September 5, the 2023 Baidu Cloud Intelligence Conference announced that on the Baidu AI Cloud Qianfan large model platform, the number of monthly live enterprises has reached nearly 10000, covering more than 400 scenes in finance, manufacturing, energy, government affairs, transportation and other industries. Huang Shi, Director Researcher of China Mobile Research Institute, introduced that China Mobile has released two industry models, namely the "Nine Days · Sea Computing Government Model" and the "Nine Days · Customer Service Model", and is continuing to deepen cooperation with market demand parties in fields such as government affairs, healthcare, justice, education, and jointly promoting the application of big model technology in practical scenarios. Behind the massive model of technological challenges to be solved and the reshaping of thousands of industries, there are numerous complex technological challenges. With the improvement of parameters and data, the demand for computing power in large models is constantly increasing, and the supply of GPU chips that provide computing power is already in short supply. Huang Shi suggests that the next step is to build a new infrastructure with computing power networks as the core, to strengthen computing with networks and alleviate the shortage of computing power supply. "It is necessary to connect ubiquitous computing resources through computing networks, use the advantages of network clusters to break the performance limit of single point computing, and alleviate the problem of insufficient computing power through the optimal configuration and use of the entire network resources." "Finding the right landing scenario is the biggest challenge faced by the big model." Xin Zhou, general manager of Baidu AI Cloud AI and big data platform, said that Baidu has studied more than 400 scenarios at present, Each scenario has different usage methods and industry experience for large models. In some benchmark industry scenarios, Baidu's algorithm experts will summarize common features for different customers and needs, continuously accumulate experience, accumulate tools, and improve the tool chain. Huawei Cloud CTO Zhang Yuxin believes that with the emergence of large model capabilities