Wind统计显示，11月25日，申万31个一级行业中有24个行业获融资净买入，其中，通信行业获融资净 买入额居首，当日净买入16.51亿元；获融资净买入居前的行业还有国防军工、电子、机械设备、计算 机、交通运输、房地产等。 （文章来源：证券时报网） 个股方面，11月25日，有1775只个股获融资净买入，净买入金额在3000万元以上的有154股。其中，25 股获融资净买入额超1亿元。新易盛获融资净买入额居首，净买入5.13亿元；融资净买入金额居前的还 有光库科技、中兴通讯、立讯精密、中际旭创、胜宏科技、长芯博创、阳光电源、易点天下等股。 ...

Core Insights - The AI industry is facing unprecedented bottlenecks as large model parameters reach trillion-level, with issues such as low efficiency of Transformer architecture, high computational costs, and disconnection from the physical world becoming increasingly prominent [2][4][38] - ZTE's recent paper, "Insights into Next-Generation AI Large Model Computing Paradigms," analyzes the core dilemmas of current AI development and outlines potential exploratory directions for the industry [2][38] Current State and Bottlenecks of LLMs - The performance of large language models (LLMs) is heavily dependent on the scaling laws, which indicate that ultimate performance is tied to computational power, parameter count, and training data volume [4][5] - Building advanced foundational models requires substantial computational resources and vast amounts of training data, leading to high sunk costs in the training process [5][6] - The efficiency of the Transformer architecture is low, with significant memory access demands, and the current hardware struggles with parallel operations in specific non-linear functions [6][7] Challenges in Achieving AGI - Current LLMs exhibit issues such as hallucinations and poor interpretability, which are often masked by the increasing capabilities driven by scaling laws [9][10] - There is ongoing debate regarding the ability of existing LLMs to truly understand the physical world, with criticisms focusing on their reliance on "brute force scaling" and lack of intrinsic learning and decision-making capabilities [9][10] Engineering Improvements and Optimizations - Various algorithmic and hardware improvements are being explored to enhance the efficiency of self-regressive LLMs, including attention mechanism optimizations and low-precision quantization techniques [12][13][14] - Innovations in cluster systems and distributed computing paradigms are being implemented to accelerate training and inference processes for large models [16][17] Future Directions in AI Model Development - The industry is exploring next-generation AI models that move beyond the Next-Token Prediction paradigm, focusing on models based on physical first principles and energy dynamics [24][26] - New computing paradigms, such as optical computing, quantum computing, and electromagnetic computing, are being investigated to overcome traditional computational limitations [29][30] ZTE's Exploration and Practices - ZTE is innovating at the micro-architecture level, utilizing advanced technologies to enhance AI accelerator efficiency and exploring new algorithms based on physical first principles [36][38] - The company is also focusing on the integration of hardware and software to create more efficient AI systems, contributing to the industry's shift towards sustainable development [38]

电th | 股票代码 | 股票 | 收盘 | 融资净买 占流通市值比例 | 涨跌幅 | | | --- | --- | --- | --- | --- | --- | | | 简称 | 入额 价 | (%) | (%) | | | | | (万元) | | | | | 300502.SZ 新易盛 304.72 | | 51287 | 0 19 | 4 | 通信 | | 300620.SZ | 光库科 | 147 34017.84 | 0.94 | 20 | 通信 | | | 技 | | | | | | 000063.SZ | 中兴道 | 39.86 33299.82 | 0.21 | 3.13 | 通信 | | | 计 | | | | | | 002475.S | 立讯精 | 55.7 32653.04 | 0.08 | 4.6 | 电子 | | | 军 | | | | | | 300308.SZ | 中际旭 | 479.66 26320.61 | 0.05 | 5 | 通信 | | | 创 | | | | | | 300476.SZ | 胜宏科 | 265.05 23923.47 | 0.11 | 5.26 ...

Company Relationships - Nvidia 与中兴通讯等公司合作，而中兴通讯因国家安全问题已在美国被有效禁止销售其产品 [1]

Company Partnerships - Nvidia 与中兴等公司合作，而中兴因国家安全问题已在美国被禁止销售其产品多年 [1]

Market Overview - The digital economy sector increased by 1.49% on November 25, with Jiarun Technology leading the gains [1] - The Shanghai Composite Index closed at 3870.02, up 0.87%, while the Shenzhen Component Index closed at 12777.31, up 1.53% [1] Top Gainers in Digital Economy Sector - Jiarun Technology (301117) closed at 48.17, up 20.00% with a trading volume of 285,900 shares and a transaction value of 1.322 billion [1] - Pinggao Co., Ltd. (688227) closed at 70.58, up 19.99% with a trading volume of 98,300 shares and a transaction value of 694 million [1] - Fuchun Co., Ltd. (300299) closed at 9.09, up 14.34% with a trading volume of 2,583,100 shares and a transaction value of 2.252 billion [1] - Lingyun Optics (688400) closed at 40.90, up 10.18% with a trading volume of 33,190 shares and a transaction value of 1.340 billion [1] - Lezhi Group (002398) closed at 6.23, up 10.07% with a trading volume of 808,900 shares and a transaction value of 493 million [1] Top Losers in Digital Economy Sector - Yingfang Software (688435) closed at 53.96, down 6.51% with a trading volume of 75,200 shares and a transaction value of 423 million [2] - ST Dongyi (002713) closed at 15.15, down 4.66% with a trading volume of 669,000 shares and a transaction value of 1.019 billion [2] - ST Tianyu (300205) closed at 4.52, down 3.62% with a trading volume of 54,700 shares and a transaction value of 25.194 million [2] Capital Flow Analysis - The digital economy sector experienced a net outflow of 2.314 billion from institutional investors, while retail investors saw a net inflow of 2.859 billion [2][3] - Major stocks like Yangguang Electric (300274) had a net inflow of 1.058 billion from institutional investors, while ZTE Corporation (000063) had a net inflow of 410 million [3] - Fuchun Co., Ltd. (300299) had a net inflow of 177 million from institutional investors, but a net outflow of 143 million from retail investors [3]

Core Viewpoint - ZTE Corporation's stock rose over 4% following the announcement of its self-developed high-performance chip "Han域" M1, which has been integrated into GAC Group's Aion brand electric vehicle model [1] Group 1: Company Developments - ZTE's subsidiary, ZTE Microelectronics, has developed the "Han域" M1 chip, which features high computing power, high bandwidth, large storage, and high security [1] - The "Han域" M1 chip is classified as an automotive-grade central computing platform SOC chip, designed to support various application scenarios in future domain-centric architectures [1] Group 2: Product Features - The chip utilizes a self-developed network acceleration engine to achieve seamless switching between three networks, enhancing its versatility [1] - It incorporates a multi-core heterogeneous system that supports cross-network forwarding and switching, along with a hardware acceleration engine for information security [1] - The on-chip storage capacity is high, and the real-time domain performance is improved by 25% compared to similar products in the international market [1]

通信行业资金流入榜 | 代码 | 简称 | 今日涨跌幅（%） | 今日换手率（%） | 主力资金流量（万元） | | --- | --- | --- | --- | --- | | 600105 | 永鼎股份 | 10.01 | 13.02 | 93525.87 | | 300308 | 中际旭创 | 5.00 | 3.97 | 68671.59 | | 300502 | 新易盛 | 4.00 | 5.24 | 65102.23 | | 300548 | 长芯博创 | 13.87 | 19.49 | 62964.93 | | 688027 | 国盾量子 | 5.03 | 6.76 | 54156.56 | | 300620 | 光库科技 | 20.00 | 19.11 | 53871.16 | | 000063 | 中兴通讯 | 3.13 | 3.75 | 32508.07 | | 688205 | 德科立 | 20.00 | 13.35 | 27218.24 | | 300570 | 太辰光 | 9.20 | 14.07 | 22885.05 | | 601869 | 长飞光纤 | 9.99 | 2.33 ...

Core Viewpoint - ZTE Corporation's stock price increased by over 4% following the announcement of its new chip, the "Han域" M1, which has been integrated into GAC Group's Aion brand electric vehicle model, the "Hao Bo GT Climbing Edition" [1] Group 1: Company Developments - ZTE Corporation's subsidiary, ZTE Microelectronics, has developed the "Han域" M1 chip, which is a high-performance, high-bandwidth, high-storage, and high-security multi-domain fusion high-speed connection chip [1] - The "Han域" M1 chip is classified as an automotive-grade central computing platform SOC chip, featuring a self-developed network acceleration engine that enables seamless switching between three networks [1] - The chip supports multi-core heterogeneous systems for cross-network forwarding and switching, enhancing real-time domain performance by 25% compared to similar products in the international market [1] Group 2: Market Impact - As of the report, ZTE Corporation's stock was trading at HKD 29.88, with a trading volume of HKD 548 million [1]

Group 1 - The top 20 stocks with significant capital inflow as of November 24 include BlueFocus (1.143 billion yuan), DataPort (806 million yuan), and 360 (794 million yuan) [1] - Other notable stocks in the top 20 by capital inflow are Provincial Advertising Group (587 million yuan), China Shipbuilding (440 million yuan), and Zhangjiang Hi-Tech (403 million yuan) [1] - Additional companies with substantial inflows include Changxin Bochuang (357 million yuan), China Shipbuilding Defense (347 million yuan), and Leike Defense (317 million yuan) [1] Group 2 - The list also features Huajian Group (286 million yuan), Aerospace Development (269 million yuan), and ZTE Corporation (259 million yuan) [1] - Other companies in the top 20 include GAC Group (253 million yuan), Changying Precision (253 million yuan), and Great Wall Military Industry (248 million yuan) [1] - The final entries in the top 20 are Zhongfu Circuit (248 million yuan), Leo Group (243 million yuan), Kunlun Wanwei (240 million yuan), Midea Group (233 million yuan), and Nanwei Software (223 million yuan) [1]