Group 1 - China's large language models are rapidly expanding into the overseas developer market, with the top ten models on OpenRouter using approximately 14 trillion tokens last week [1] - MiniMax M 2.5 leads with 2.57 trillion tokens, followed by Kimi K2.5 with 1.04 trillion tokens, indicating widespread adoption of Chinese AI models by overseas tech companies [1] - Hong Kong-listed companies, Zhipu and MiniMax-WP, have seen cumulative gains exceeding 400% since their listings, establishing themselves as the recognized leaders in AI models [1] Group 2 - Morgan Stanley recently initiated coverage on Zhipu and MiniMax, assigning an "overweight" rating, signaling a strategic recognition of Chinese AI model firms by international capital [1] - The Hong Kong market is acting as a "buffer zone" amid the ongoing US-China tech rivalry, becoming a crucial node for Chinese AI companies to connect with global capital [1] Group 3 - By 2026, the Hong Kong Stock Exchange is witnessing a surge in listings from Chinese AI companies, including GPU firms like Birran Technology and Tensu Zhixin, filling gaps in core computing power [2] - Over 40 AI-related companies are now listed on the Hong Kong Stock Exchange, categorized into four main sectors: AI computing power industry chain, AI software (including large models), AI hardware (covering mobile phones, cars, robots, etc.), and large internet enterprises [2] - Driven by policy support and market demand, related concept stocks are entering a new growth cycle [2]

Market Overview - The Hong Kong stock market experienced a decline across all major indices, with the Hang Seng Index falling by 1.82% to close at 26,590.32 points, the Hang Seng China Enterprises Index down by 2.06% to 9,007.86 points, and the Hang Seng Tech Index decreasing by 2.13% to 5,270.70 points due to concerns over tariffs and geopolitical tensions [1][2]. Sector Performance - Large technology stocks saw a significant drop, with Meituan down over 4%, and other major players like Tencent, Xiaomi, Alibaba, and JD.com also closing lower [2]. - Financial stocks were weak, with China Life Insurance dropping by 6% and New China Life Insurance down by 5.79% [2]. - The film sector faced substantial pressure, with notable declines in stocks such as Maoyan Entertainment, which fell over 8%, and Huayi Brothers down over 5% [2][7]. AI and Technology Stocks - AI application stocks showed strong performance, with Zhizhu rising by over 12% and Zhaoyi Innovation increasing by nearly 12% as well [5][6]. - Other notable gainers in the AI sector included MINIMAX-WP, which rose by over 4%, and companies like Huahong Semiconductor and Lenovo also saw positive movement [5][6]. Film Industry Insights - The 2026 Spring Festival box office reported a total of 5.752 billion yuan, marking a year-on-year decline of approximately 40%, with attendance dropping by 36.2% compared to 2025 [9]. - The average ticket price saw a significant reduction of nearly 3 yuan from the previous year's 50.8 yuan, indicating a need for improved content quality in the film industry [9].

Market Overview - The Hong Kong stock market experienced a decline, with the Hang Seng Index falling by 1.93% to 26,560.57 points, the Hang Seng Tech Index down by 2.36% to 5,258.33 points, and the Hang Seng China Enterprises Index decreasing by 2.06% to 9,007.78 points [1][2]. Major Stocks Performance - Major technology stocks saw significant declines: Alibaba fell by 2.96%, Tencent by 3.44%, JD.com by 0.19%, Xiaomi by 1.86%, NetEase by 0.71%, Meituan by 4.76%, Kuaishou by 3.79%, and Bilibili by 2.73% [2]. Film and Entertainment Sector - The film and entertainment sector faced substantial drops, with DMG Entertainment down by 9% and Maoyan Entertainment falling over 7%, hitting new lows. According to a report from CICC, the effective box office for the first six days of the 2026 Spring Festival was 5.077 billion yuan, a decrease of 38.7% compared to the previous year. The expected box office for the period may fall below the previously anticipated range of 6.5 to 8.5 billion yuan due to a lack of leading films [3]. Tourism and Retail Sector - The tourism and retail sector also declined, with China Duty Free Group dropping over 8%. During the first five days of the Spring Festival holiday, duty-free shopping in Hainan reached 1.38 billion yuan, with 177,000 shoppers, representing increases of 19% and 24.6% year-on-year, respectively. However, the stock price of China Duty Free has largely reflected this demand [3]. Semiconductor Sector - The semiconductor and memory stocks showed resilience, with Hynix reporting strong demand from AI clients and limited supply, leading to continued price increases in memory products. Hynix indicated that its inventory for DRAM and NAND is only about four weeks, and it expects this level to continue to decline throughout the year [4]. AI Applications - AI application companies, particularly Zhiyuan and MINIMAX-WP, performed strongly, with Zhiyuan rising nearly 21% and MINIMAX-WP up nearly 8%. Reports indicate that Chinese large models are capturing a significant share of the global developer market, with Chinese models accounting for 61% of the total token volume [4].

Group 1 - The core viewpoint of the article highlights the strong performance of major AI model companies, with Zhiyu (02513) rising by 9.82% to 615 HKD and MINIMAX-WP (00100) increasing by 7.73% to 905.5 HKD [1] - OpenRouter's latest weekly data shows that the total token volume of the top ten models is approximately 8.7 trillion, with Chinese models dominating at 5.3 trillion, accounting for 61% [1] - MiniMax M2.5 ranks first with 2.45 trillion tokens, followed by Kimi K2.5 with 1.21 trillion tokens, while Zhiyu GLM 5 and DeepSeek V3.2 rank third and fifth respectively [1] Group 2 - Guolian Minsheng Securities believes that the trend in the AI industry has not changed this year, transitioning from technological research and development to large-scale commercial implementation [1] - In the long term, the continuous iteration of reasoning and multimodal large models, integration of major ecosystems, and increasing penetration across various industries are expected to drive rapid growth in token usage [1] - The large-scale commercial application and industrial implementation of large models are anticipated to be promising [1]

Summary of Conference Call on AI Industry and Company Developments Industry Overview - The release of Zhiyu GLM 5 marks a significant enhancement in domestic AI models' capabilities in coding and agent tasks, transitioning from code completion to task execution, laying the foundation for commercial applications [2][3] - The domestic AI industry is expected to enter a new phase in the second half of 2025, with models like GLM, MinMax, and kimi gaining global recognition due to their competitive pricing, which remains at one-fifth of similar overseas services [2][5] Key Points on Zhiyu GLM 5 - Zhiyu GLM 5's success is attributed to its strong technical support from Tsinghua University, a high talent density, and active support for domestic computing power [2][6] - The model excels in complex tasks such as programming and light rendering, making it a competitive AI model against top international models like OPPO 4.7 and OPPO 4.6 [6] Pricing and Market Dynamics - Despite recent price increases, domestic AI models remain significantly cheaper than their overseas counterparts, providing a competitive advantage in the global market [4][9] - API and subscription prices for Zhiyu GLM 5 have risen significantly, with basic input prices set at 4 RMB and 6 RMB, and output prices increasing by 125% to 250% for basic token usage [8] Future Trends - The integration of domestic AI models with office tools is expected to become more common starting in 2026, exemplified by the launch of GLM in Excel, which will transform traditional data collection and presentation processes [7] - The demand for cloud computing resources is anticipated to rise alongside the increasing need for AI applications, leading to a significant growth in cloud service demand and revenue for companies like Zhiyu [10] Investment Opportunities - Key areas to watch in the coming years include AI models (Zhiyu, MinMax, Alibaba, ByteDance, Tencent), agent applications (e.g., TaxFriend, Zhongkong, Hehe, Kingsoft), and cloud computing (e.g., Alibaba Cloud, Capital Online, Kingsoft Cloud, Deepin) [11]

2026-02-12 摘要 智谱 GLM 5 在长期任务中取得显著进步，缩小了与 Cloud OPIUS 4.5 的差距，并在部分 benchmark 测试中超越后者。GLM 5 的 API 和 Coding Plan 已上线，API 定价已公布，Coding Plan 将于 2 月 16 日 正式上线。 智谱推出了特工模式（视角转换、数据分析、AI PPT 支持）、数据洞察 （数据解析、画图、导出多种格式文件）和写作功能（支持多种写入模 式并导出 PDF 和 Word）。 Minimax 发布了 M2.5，是首个为 Agent 场景原生设计的生产级模型， 支持 Excel 处理、深度调研及 PPT 等生产力场景，但 API 尚未上线， 也未公布价格。 GLM 5 的 Output 定价为每百万 Tokens $3.2，GLM 4.6 和 GLM 4.7 从 $1.5 提升至 $2.2，国内 API 价格提升幅度在 33%至 100%之 间，海外市场为 67%至 100%。Coding Plan 国内 Light 和 Max 两个 季度收费版本分别提高了 23%和 18%；海外市场分别提高了 60%和 30% ...

Summary of Key Points from the Conference Call Industry Overview - The AI video production industry is transitioning from high-cost, long-cycle, and high-threshold production to low-cost, efficient, and democratized industrial production, reshaping supply and demand dynamics and creating investment opportunities. However, there are concerns about the potential impact of large model capabilities on applications [2][3]. Core Insights and Arguments - The announcement of price increases by cloud service providers like Amazon AWS indicates the growing scarcity of AI cloud infrastructure, confirming the high demand for global AI computing power. The market is expected to replicate the post-Chinese New Year rally seen in 2025 [2][3]. - The release of models such as CloudOps 4.6 and Zhizhu GLM 5 marks a shift in AI from chatbots to customized team collaboration executors, which is anticipated to drive exponential growth in the industry by 2026 and change digital work patterns for businesses and individuals [2][5]. - Investment opportunities in domestic computing power related to AI applications are promising, with recommendations for GPU companies like Cambricon and self-developed ASIC chips from internet firms. Attention is also drawn to the expected price increases in target materials, with companies like Jiangfeng Electronics, OLAI New Materials, and Ashi Chuang being highlighted [2][6]. Investment Opportunities - The price increase of CCL (copper foil substrate) is expected to manifest in Q1 2026, with a projected overall positive performance for 2026-2027. Beneficiary companies include Nanya New Materials, Shengyi Technology, and Huazheng New Materials [2][8]. - In the AI hardware sector, there are significant investment opportunities related to domestic computing power, particularly due to the high power consumption of various AI applications, especially video-related ones [6][7]. Military Applications of AI - AI is primarily applied in military fields for situational awareness and unmanned operations, with Palantir's $4.3 billion large order demonstrating the substantial commercial potential of AI in military situational awareness [4][10][11]. - The development of unmanned operations is expected to accelerate significantly, with a focus on increasing the proportion of unmanned equipment in military operations [12][13]. Communication Chain Opportunities - Opportunities in the AI industry chain within the communication sector are identified in three main areas: AIS cloud services, network endpoints, and computing endpoints. Price increases in related components are anticipated due to the growth of AI applications and model upgrades [14]. Media and Entertainment Sector - The media and internet industry shows significant investment opportunities in AI applications, particularly with ByteDance's CDS 2.0 model, which enhances video production capabilities and efficiency. Companies with strong IP and distribution capabilities are recommended for investment [15]. Gaming Industry Impact - AI technology is expected to significantly enhance game development efficiency and user experience. Companies like Perfect World and KAEYING Network are highlighted as key investment targets due to their innovative projects [16][17]. Pharmaceutical Industry Potential - AI has substantial potential in the pharmaceutical industry, particularly in drug development, where it can shorten development cycles and improve efficiency. Companies like Hongbo Pharmaceutical and others are noted for their advancements in utilizing AI for drug development [18]. Multimodal Model Development - The development of multimodal models is rapidly advancing, with significant improvements in capabilities and usability. Companies like ByteDance and others are leading in this space, indicating a competitive landscape similar to the early stages of text models [19][20][21][22]. Recommended Companies for Investment - In the multimodal direction, companies such as Wanjing Technology, Haitai Ruisheng, and others are recommended for their strong positions in AI creative tools and applications [26]. This summary encapsulates the key insights and investment opportunities discussed in the conference call, providing a comprehensive overview of the current landscape in the AI industry and its various applications across sectors.