证券研究报告|行业研究周报 [Table_Date] 2026 年 2 月 1 日 算力供给紧张，成为 2026 年主线 [Table_Title2] 计算机行业周报 [Table_Summary] 本周观点: 算力供给紧张，成为 2026 主线 全球 AI 基建需求依然强劲，26 年海外 AI 算力依然是主要配置方 向。TrendForce 预测 2025 年全球八大主要 CSPs 资本支出(CapEx) 总额约 4,300 亿美元(yoy+65%),预期 2026 年 CSPs 仍将维持积极 的投资，资本支出将进一步推升至 6,000 亿美元以上(yoy+40%), 展现出 AI 基础建设的长期成长潜能。 海外大厂创造 2026 算力新格局 根据财经宇宙，微软 CEO Satya Nadella 表示未来微软从 OpenAI 的"独家运营商"，转型为全球 AI 资源的总调度商，让企业客户 能根据成本与性能需求，在 GPT 5.2、Claude4.5 等多款大模型间 灵活选择。微软正通过这种饱和式的基建投入，试图在 AI 从技术 神话下沉为工业底座的过程中，完成从一家"软件服务商"向未来 "算力基础服务商" ...

Core Viewpoint - Molica AI offers a comprehensive platform integrating multiple advanced AI tools at a significantly reduced price, promoting accessibility and usability for a wide range of users [1][2][10]. Group 1: Product Offering - Molica AI aggregates over 20 leading AI models, including ChatGPT5.2, Gemini3.0, Nano Banana Pro, Sora2, Claude4.5, and others, providing users with a one-stop solution for various AI needs [2][12]. - The platform has been operational for three years and boasts over 1 million members, indicating a stable and growing user base [2][6]. - Users can access all integrated AI tools for a one-time fee of 298 yuan, down from the original price of 998 yuan, with a limited offer for the first 100 members [4][69]. Group 2: Key Features of Integrated AI Tools - Gemini3.0 Pro, ChatGPT5.2, and Claude4.5 are highlighted as top-tier AI models, excelling in reasoning, coding, and content generation [7]. - Nano Banana Pro is noted for its superior text rendering capabilities, especially in Chinese, and can produce 4K images with customizable dimensions [15][34]. - Sora2 and Veo3.1 are recognized for their exceptional video generation capabilities, allowing users to create high-quality marketing content effortlessly [36][41]. - Midjourney remains a leader in AI art generation, providing high-quality images and creative outputs that rival top designers [43][49]. - SUNO enables users to generate complete songs from simple prompts, democratizing music creation [54][57]. Group 3: Membership Benefits - Membership includes access to a comprehensive AI monetization manual valued at 2000 yuan, which provides real-world applications and strategies for leveraging AI in various industries [61][63]. - The platform emphasizes continuous updates to the manual, ensuring users stay informed about the latest AI trends and monetization methods [61][64]. - Users can learn advanced AI operational techniques and practical applications for side projects, enhancing productivity across different sectors [62][63].

Core Viewpoint - Molica AI offers a year-end promotion for a lifetime membership at a significantly reduced price, aggregating multiple advanced AI tools into one platform, making it accessible and cost-effective for users [2][70]. Group 1: Promotion Details - The year-end promotion offers a lifetime membership originally priced at 998 yuan for only 296 yuan, limited to the first 100 users [2][70]. - The promotion runs from December 24 to December 26, providing access to over 20 AI tools including ChatGPT5.2, Gemini3.0, and others [2][68]. Group 2: AI Tools Overview - Molica AI integrates leading AI models such as Gemini3.0 Pro, ChatGPT5.2, and Claude4.5, which excel in various tasks like reasoning, coding, and content generation [6][11]. - Nano Banana Pro, based on Gemini 3.0, enhances text rendering capabilities, especially in Chinese, and supports 4K output with customizable image dimensions [14][31]. - Sora2 and Veo3.1 are highlighted for their video generation capabilities, allowing users to create high-quality marketing content easily [35][40]. - Midjourney is recognized for its superior image quality and creativity in AI-generated art, maintaining a strong position in the AI art space [43][49]. - SUNO enables users to generate complete songs from simple prompts, democratizing music creation [52][56]. Group 3: Membership Benefits - Membership includes access to all integrated AI tools and features, providing a comprehensive solution for various creative and professional needs [68][69]. - Users can benefit from additional resources such as an AI monetization manual, which offers practical guidance on leveraging AI for side projects [60][61]. Group 4: Market Positioning - Molica AI positions itself as a cost-effective alternative to individual AI subscriptions, which can be significantly more expensive when purchased separately [67]. - The company emphasizes the importance of AI in modern work environments, suggesting that those who effectively utilize AI will have a competitive advantage [66].

Core Insights - Google Gemini 3 Pro shows significant advancements over its predecessor, Gemini 2.5 Pro, outperforming GPT-5.1 and Claude 4.5 in nearly all benchmark tests, including academic reasoning and visual reasoning puzzles [1][2]. Benchmark Performance - In "Humanity's Last Exam," Gemini 3 Pro scored 37.5% without tools and 45.8% with search and code execution, compared to 21.6% for Gemini 2.5 Pro [2]. - For the ARC-AGI-2 visual reasoning puzzles, Gemini 3 Pro achieved 31.1%, a substantial increase from 4.9% in Gemini 2.5 Pro [2]. - In mathematics, Gemini 3 Pro scored 95.0% in AIME 2025 without tools and achieved a perfect score of 100% with code execution [2]. - The LiveCodeBench Pro benchmark saw Gemini 3 Pro with an Elo Rating of 2,439, significantly higher than Gemini 2.5 Pro's 1,775 [2]. Model Evolution - The Gemini series has evolved significantly, with each generation addressing the shortcomings of the previous one. The first generation established multimodal capabilities, while the second focused on decision-making and planning [15][18]. - Gemini 2.5 introduced a reasoning engine for deeper reasoning and problem-solving, leading to the current generation, which integrates multimodal, reasoning, and agent capabilities [19][20]. User Interaction and Usability - Gemini 3 Pro is designed to understand user intent better, allowing for more straightforward interactions without the need for complex prompts [21]. - The model can seamlessly process text, images, videos, audio, and code, enhancing its usability across various applications [23]. Development Platform - Google introduced the Antigravity platform alongside Gemini 3 Pro, aimed at simplifying the development process for AI agents, allowing developers to focus on higher-level tasks [29][33]. - Antigravity supports multiple models, including third-party options, and has attracted significant developer interest due to its generous rate limits [33]. Future Developments - A more advanced version, Gemini 3 Deep Think, is in development, promising further enhancements in capabilities [13][14].