以下文章来源于CSDN ，作者万有引力 CSDN . 成就一亿技术人 作者 | 万有引力 出品 | CSDN（ID：CSDNnews） 在过去短短的几周里，大模型赛道的信息密度飙升至前所未有的高度。DeepSeek 连续 五天开源 ，直接引发了一场复现热潮；阿里巴巴通义实验室、 腾讯相继推出面向视觉文档的 RAG 系统 ViDoRAG、新一代混元快思考模型 Turbo S ，加速了大模型的演进步伐；马斯克用 20 万张 GPU 训练出的 Grok 3 ，超越了许多业界标杆，再次验证了"大力出奇迹"的定律； Claude 3.7 Sonnet 迎来编码能力大升级，AI 编程的技术平权时代正在加速到来； DeepSeek 论文与 Kimi"撞车"，越来越多公司开始布局稀疏注意力与线性注意力机制，这些技术正成为 Transformer 之后的关键探索方向；此外， Manus 模式的"虚拟机"概 念迅速走红，正在重塑大模型的运行方式... 在这场眼花缭乱的技术竞赛背后，真正值得我们关注的是什么？DeepSeek 的五连发 究竟意欲何为？在 545% 的成本利润率之下，其他大模型公司是 否也能找到盈利空间？面对行业变 ...

Core Viewpoint - The article discusses the significant upgrades in the DeepSeek V3-0324 model, which, despite being labeled as a "minor version upgrade," shows substantial improvements in performance metrics compared to its predecessor and other models in the market [2][6][9]. Summary by Sections Model Performance - DeepSeek V3-0324 has demonstrated a considerable jump in all metrics during internal benchmarking, now being recognized as the best non-reasoning model, surpassing Sonnet 3.5 [6][9]. - The model achieved a perfect score in coding tests, indicating high consistency and reliability in performance [11]. Problem-Solving Capabilities - The model exhibits a unique ability to re-evaluate problems when it encounters difficulties, showcasing a form of autonomous thinking [12][13]. - An example problem involving a 7-meter long cane passing through a 2-meter high and 1-meter wide door illustrates the model's capacity to rethink and approach problems from different angles, leading to a correct solution despite initial misunderstandings [18][22]. Accessibility and Open Source - DeepSeek V3-0324 remains free and open-source, with its weight files available on HuggingFace under a permissive MIT license, maintaining the same storage requirements as its predecessor [28][29]. - The model can be accessed through various platforms, including the official website and HuggingFace, allowing users to experience its capabilities firsthand [30][32].

Core Insights - China is experiencing a significant shift towards open-source AI models, which is enhancing AI adoption and innovation, likened to an 'Android moment' for the sector [1][22] Open-Source AI Models - The open-source movement is led by AI startup DeepSeek, whose R1 model has challenged American tech dominance and raised questions about the spending of Big Tech on large language models [2][3] - DeepSeek's R1 model is distributed under an 'MIT License', allowing unrestricted use, modification, and distribution, which is seen as a catalyst for the adoption of open-source AI models in China [8][15] - Major Chinese tech companies like Baidu, Alibaba, and Tencent are increasingly offering their AI models for free and moving towards open-source strategies [12][20] Baidu's Strategy - Baidu has released its latest AI model, Ernie 4.5, and plans to make it open-source by the end of June, marking a strategic shift from its previous proprietary model [4][5] - This move is indicative of a broader trend in China, where companies are compelled to adopt open-source models to remain competitive against disruptors like DeepSeek [15][20] Competitive Landscape - The emergence of DeepSeek has pressured other Chinese competitors to adopt open-source business models, as they cannot charge for similar offerings that are available for free [15][21] - OpenAI and other U.S. companies continue to operate under a proprietary model, raising questions about their pricing strategies in light of the competitive open-source landscape [16][20] Market Dynamics - The open-source trend is expected to drive down costs and foster innovation, with Chinese companies historically excelling in product innovation [21][22] - Experts suggest that the rapid adoption of open-source models in China could narrow the technological gap with the U.S., previously estimated at 12 to 24 months [22][23]

Core Viewpoint - The tech-heavy Nasdaq Composite has declined over 8% this year, prompting investors to seek opportunities, particularly in popular stocks like Nvidia and Tesla, which have seen significant price drops [1][2]. Group 1: Tesla - Tesla's stock has dropped 38% this year, influenced by broader market sell-offs and concerns regarding first-quarter deliveries [2][3]. - The stock is viewed as a battleground, with analysts divided on its valuation and future prospects; it trades at approximately 86 times forward earnings, with expected earnings growth of 17% [4]. - Among 36 analysts, 12 recommend buying Tesla, 13 suggest holding, and 11 advise selling, with an average price target indicating a potential upside of 41% [4]. - Concerns have been raised about CEO Elon Musk's political involvement and its potential negative impact on Tesla's sales [5]. - Investors are closely monitoring Tesla's first-quarter earnings, as high valuations could lead to significant stock price volatility based on performance [6]. Group 2: Nvidia - Nvidia is widely regarded as a strong buy, with 39 out of 42 analysts recommending the stock, suggesting nearly 50% upside potential [7]. - The emergence of a Chinese start-up, DeepSeek, which developed a chatbot using older Nvidia chips, raised questions about the necessity of Nvidia's technology in the AI sector [8]. - Despite these concerns, Nvidia's CEO Jensen Huang remains optimistic, asserting that increased data demand will enhance the need for Nvidia's products [9]. - Nvidia's stock now trades at 26 times forward earnings, down from previous peaks above 50, with analysts projecting earnings growth of about 48% for fiscal year 2026 [10].

Core Viewpoint - Achieving "steady growth" in China requires a combination of macroeconomic, industry, and institutional policy measures, with a gradual shift from macro policies to industry and reform policies for sustainable growth [1][11]. Macroeconomic Policies - The 2025 government work report sets a GDP growth target of around 5%, a CPI increase of about 2%, and a fiscal deficit rate raised to 4%, indicating a more proactive fiscal policy stance and continued moderate monetary policy [3]. - The focus on total factor productivity improvement is crucial for maintaining steady growth, emphasizing the need for continuous enhancement of enterprise competitiveness and rapid industrial upgrades to avoid falling into "middle-income traps" [3][5]. Industry Policies - The report highlights the dual approach of developing new productive forces and upgrading traditional industries, with industry upgrades becoming a key policy focus after stabilizing overall demand [3][11]. - Innovation is identified as the key to overcoming challenges posed by rising costs, demographic changes, and international market conditions, necessitating a shift from factor-driven growth to innovation-driven growth [5][6]. Innovation Capability - The need to enhance innovation capability is underscored, as past growth relied on low-cost advantages, while future growth must depend on original inventions and discoveries [6][8]. - China is positioned to leverage the Fourth Industrial Revolution, with significant advancements in digital economy and artificial intelligence, providing opportunities for sustained economic growth [6]. Role of Private Enterprises - Private enterprises contribute over 70% to China's innovation, making their support essential for fostering innovation [9]. - Recent government policies aim to bolster private enterprises' confidence and participation in major national projects, emphasizing the importance of effective implementation of these policies [9][12]. Sustainable Growth Policies - A combination of macro, industry, and reform policies is necessary for achieving sustainable growth, with a focus on addressing structural risks in key industries and nurturing new growth drivers [11][12]. - The government should create a conducive environment for innovation, particularly by supporting private enterprises and ensuring effective market and government roles [12].

Core Viewpoint - Alibaba is undergoing a significant transformation towards AI, driven by Jack Ma's return and a strategic shift under new CEO Wu Yongming, with plans to invest 380 billion yuan in AI infrastructure over the next three years, surpassing the total investment of the past decade [1][14][15]. Group 1: Strategic Shift and Investment - Jack Ma emphasizes the importance of AI for Alibaba's future, leading to a comprehensive transformation strategy [1][2]. - The company plans to invest 380 billion yuan in AI infrastructure, which is more than the total investment in the past ten years [14][15]. - Alibaba's stock has risen 66% since the beginning of the year, reflecting investor optimism about the strategic shift [3]. Group 2: Competitive Landscape - The launch of ChatGPT by OpenAI highlighted Alibaba's lag in AI, prompting a strategic pivot to regain competitive advantage [2]. - Alibaba's Qwen language model has established a leading position in the Chinese market, even gaining recognition from Apple for its AI functionalities on iPhones [2][13]. - The rise of local startups like DeepSeek has intensified competition, showcasing the potential of Chinese AI firms [2][14]. Group 3: Leadership and Organizational Changes - Wu Yongming has been appointed as CEO, focusing on consolidating decision-making and reallocating resources towards AI [10][11]. - The company has undergone a restructuring plan to split into six independent units, although this was later adjusted due to market conditions [6][8]. - Jack Ma's return has reinvigorated the company's direction and morale, with a renewed focus on AI as a core belief [3][17]. Group 4: Future Prospects and Challenges - Alibaba aims to achieve Artificial General Intelligence (AGI), which has boosted internal morale despite skepticism from experts [17]. - The company is facing challenges from competitors like Tencent and ByteDance, who are rapidly advancing their AI capabilities [18][19]. - Geopolitical tensions and restrictions on advanced AI chip exports may hinder Alibaba's ambitions in AI development [18][19].

Core Viewpoint - The U.S. stock market is under pressure, but the ADRs of Chinese stocks, particularly Alibaba, are gaining traction with significant potential for further upside [1] Group 1: AI Leadership - Alibaba is a leader in artificial intelligence (AI), with its Qwen 2.5 model outperforming competitors including DeepSeek and U.S. firms like Meta Platforms and OpenAI [2] - The company has launched over 100 task-specific open-source AI models, including those for mathematics and coding, and introduced a new AI assistant powered by its QwQ-32B AI reasoning model [3] - Revenue from Alibaba's Cloud Intelligence segment grew 13% last quarter, with AI-related revenue more than doubling and segment-adjusted EBITDA increasing by 33% [4] - Partnerships with major tech companies, such as Apple using Alibaba's AI model for its Apple Intelligence solution in China, highlight Alibaba's growing influence in the AI space [5] Group 2: E-commerce Recovery - Alibaba is showing signs of recovery in its core e-commerce business, which includes Tmall and Taobao, after facing challenges from a sluggish Chinese economy and competition [6][7] - Investments in the e-commerce segment have led to a 9% increase in third-party revenue and a 5% rise in overall segment revenue last quarter, with segment EBITDA up by 2% [8] Group 3: Emerging Business Growth - Alibaba's International commerce segment (AIDC) is expanding rapidly, with a 32% revenue increase last quarter, although it currently has a negative EBITDA of $678 million [9][10] - Management anticipates that the AIDC segment will achieve profitability within the next fiscal year, which would significantly enhance the company's earnings growth [10] Group 4: Stock Valuation - Despite a 60% increase in share price year-to-date, Alibaba's stock is still attractively valued, trading at a forward P/E ratio of about 15 for fiscal 2026, which is approximately half that of Amazon [11][12] - The company holds $23.1 billion in cash and short-term investments, along with $47.4 billion in equity and other investments, representing over 20% of its market cap [12] - There is potential for Alibaba to accelerate revenue and earnings growth, making it a compelling investment opportunity [13]

Core Viewpoint - The article discusses the challenges faced by Nvidia following its GTC conference, highlighting investor concerns over tariffs, competition from emerging models like DeepSeek, and the trend of major tech companies developing their own AI chips, which threaten Nvidia's market dominance [1][2][6]. Group 1: Market Reaction - Despite the unveiling of new products at the GTC conference, Nvidia's stock fell by approximately 4% after CEO Jensen Huang's keynote, contributing to a nearly 12% decline year-to-date [1][6]. - The conference attracted a record 25,000 attendees, indicating strong interest in AI technology, yet this enthusiasm did not translate into positive market sentiment for Nvidia [3]. Group 2: Political and Economic Threats - Nvidia faces significant threats from political factors, particularly Trump's tariff policies, which could impact its global operations. Huang attempted to reassure investors that tariffs would not cause "significant damage" in the short term, but did not guarantee long-term immunity from economic impacts [8][9]. - The company plans to invest billions in U.S. manufacturing to diversify its supply chain, which may increase operational costs and affect profit margins [9]. Group 3: Competitive Landscape - The rise of cost-effective models like DeepSeek poses a threat to the demand for high-end AI chips, as the AI industry may not require Nvidia's expensive chips if alternatives become viable [11]. - Huang argued that the demand for computing remains extremely high, driven by AI inference, and that new AI models will increase the need for robust computing infrastructure [12]. Group 4: Client Self-Development - Major tech companies, including Meta, Google, Amazon, and Microsoft, are actively developing their own AI chips to reduce reliance on Nvidia, which could undermine Nvidia's monopoly in the AI chip market [13]. - Reports indicate that Amazon Web Services (AWS) is promoting its self-developed Trainium chips, which offer similar performance to Nvidia's H100 chips at a quarter of the price, further intensifying competitive pressure on Nvidia [13].

Core Viewpoint - The article discusses the rapid growth and impact of the DeepSeek AI model, highlighting its adoption by various companies and the resulting surge in demand for AI applications across industries [3][4][5]. Group 1: DeepSeek Model Adoption - The DeepSeek-R1 model, with parameters reaching 671 billion, has become a focal point for many small and medium enterprises seeking to deploy AI solutions [2]. - The launch of a verification tool for identifying the "full version" of DeepSeek has led to an unexpected increase in traffic and user engagement [3]. - Over 200 companies have integrated DeepSeek technology into their operations, indicating a significant trend in the AI industry towards rapid adoption and deployment [3][4]. Group 2: Market Dynamics and Competitive Landscape - The introduction of DeepSeek has sparked a new wave of competition in the AI sector, with companies racing to integrate and innovate using the model [3][4]. - The model's popularity has resulted in increased user demand, with DeepSeek becoming one of the fastest-growing AI applications, reaching 157 million active users in February [3]. - The competitive pricing of DeepSeek's API has led to a price war among major AI players, significantly lowering costs for businesses [14][15]. Group 3: Technological Innovations - DeepSeek's V2 model introduced significant innovations, including a multi-head latent attention architecture that reduces memory bandwidth requirements, enhancing performance [9]. - The model's cost-effectiveness has been highlighted, with a theoretical profit margin of 545%, making it attractive for both B2B and B2C applications [16][18]. - The advancements in AI technology are expected to lead to a rapid explosion of AI applications, driven by decreasing operational costs and increasing capabilities [18][19]. Group 4: Future Outlook - The article emphasizes the potential for a hybrid AI deployment model, combining cloud and edge computing to optimize performance and data privacy [20]. - As more companies integrate DeepSeek, the ecosystem is expected to evolve, fostering innovation and differentiation in AI applications [19][20]. - The ongoing development in AI infrastructure and software is laying a solid foundation for widespread AI application adoption in various sectors [20].

一对兄妹撕开了AI融资潮的一角。 近日，Ant h r o p i c宣布完成E轮35亿美元（约合人民币250亿元）融资，投后估值6 1 5亿 美元（约合人民币4 4 5 0亿元），这距离Ant hr opi c上一轮10亿美元的融资仅仅过去不到 1 0天。 达里奥·阿莫迪和丹妮拉·阿莫迪兄妹——2021年离开Ope nAI开始创业，以他们为代表的 AI新贵们正在进行一场史无前例的融资竞赛。2 0 2 5年短短3个月，从Op e nAI到xAI、再 到Ant hr op i c，以及Ope nAI前CTO创立的新AI公司都在疯狂融资。 夸张的是，不少创立时间甚至没有超过一年的年轻企业，任何一轮融资拿出来都可以造 出一只超级独角兽。 AI融资竞赛。 作者 I 杨文静 报道 I 投资界PEdaily 两兄妹联手，估值4000亿 AI世界永远有讲不完的疯狂故事。 An t hr opi c此次融资的领投方是光速创投，跟投方包括Be ss eme r Ve n t u r e Pa rt n e rs，思 科投资，D1 Ca p it a l Pa rt n e rs，富达投资，Ge ne r a l Ca t a ...