Core Viewpoint - The stock price of Jing Sheng Mechanical & Electrical (300316) surged over 15% on September 26, reaching a new high for the year, following the announcement of the successful commissioning of its first 12-inch silicon carbide substrate processing pilot line at its subsidiary, Zhejiang Jing Rui SuperSiC, marking a significant advancement in domestic technology [1][2]. Group 1: Company Developments - On September 26, Jing Sheng Mechanical & Electrical announced the successful commissioning of its first 12-inch silicon carbide substrate processing pilot line, achieving 100% domestic production and marking a transition from following to leading in global SiC substrate technology [2]. - The successful commissioning of the pilot line signifies a new phase of efficient intelligent manufacturing for the company [2]. Group 2: Industry Context - The development of AI technology has led to increased performance of GPU chips used in AI servers, resulting in higher power consumption and necessitating advanced packaging techniques to manage heat dissipation [4]. - Traditional ceramic substrates have thermal conductivity of approximately 200 W/mK to 230 W/mK, which is insufficient for the growing heat dissipation demands [4]. - Silicon carbide materials exhibit superior thermal conductivity, reaching up to 400 W/mK and nearly 500 W/mK, making them ideal for packaging in data centers and high-performance AI chips [4].

Core Viewpoint - The DeepSeek-R1 inference model research paper, led by Liang Wenfeng, has been published in the prestigious journal Nature, marking a significant milestone in the field of AI and large language models [1][3]. Group 1: Model Development and Validation - The latest paper provides more detailed insights into the training of the DeepSeek-R1 model compared to its initial version released in January [3]. - DeepSeek-R1 is recognized as the first mainstream large language model to undergo peer review, addressing previous concerns regarding its distillation process [3]. - The peer review process is seen as a necessary step to mitigate the risks associated with unverified claims in the AI industry, as highlighted by Nature [5]. Group 2: Data and Safety Assessment - DeepSeek-V3 Base, the foundational model for DeepSeek-R1, utilized data sourced entirely from the internet, which may include outputs generated by GPT-4, though this was not intentional [5]. - The company has provided a detailed process in supplementary materials to demonstrate how data contamination was minimized during training, ensuring that benchmark tests were not deliberately included to enhance model performance [5]. - A comprehensive safety assessment of DeepSeek-R1 has been conducted, showing that its safety features are superior to those of contemporaneous models [5].