Core Insights - DeepSeek has officially released the DeepSeek-V3.2-Exp model, which introduces a Sparse Attention mechanism for improved training and inference efficiency on long texts [1] - The official app, web version, and mini-program have all been updated to DeepSeek-V3.2-Exp, with a significant reduction in API costs by over 50% for developers [1] - The company has also adapted to the latest DeepSeek model and open-sourced the vLLM-MLU inference engine code, allowing developers to experience the new model on their platform [1] Model Iteration and Features - DeepSeek is progressing with model iterations, having recently upgraded to DeepSeek-V3.1-Terminus, which features a hybrid inference architecture supporting both thinking and non-thinking modes [2] - The V3.1 model boasts higher thinking efficiency and enhanced agent capabilities, showing significant improvements in tool usage and intelligent task performance [2] - The V3.1 model utilizes UE8M0 FP8 Scale parameter precision, designed for the upcoming generation of domestic chips, which has positively impacted the stock prices of related domestic chip industry companies [2]

Core Insights - DeepSeek has released the experimental version DeepSeek-V3.2-Exp, which introduces Sparse Attention for improved training and inference efficiency on long texts [1] - The API pricing has been reduced by over 50% due to significant cost savings from the new model [1] - Cambricon has adapted to DeepSeek-V3.2-Exp and open-sourced the vLLM-MLU inference engine, allowing developers to experience the new model on their platform [1][2] - Huawei Ascend has also quickly adapted to DeepSeek-V3.2-Exp, open-sourcing all inference code and achieving optimized deployment on the CANN platform [3] Group 1 - DeepSeek-V3.2-Exp is an experimental version that builds on the previous V3.1-Terminus, focusing on optimizing long text processing [1] - The new model's API pricing reduction is a strategic move to enhance developer engagement and usage [1] - Cambricon's rapid adaptation to the new model demonstrates its commitment to software ecosystem development and performance optimization [2] Group 2 - Huawei's deployment of DeepSeek-V3.2-Exp BF16 model showcases its capability in handling large sequence processing with low latency and high throughput [3] - The continuous iteration of DeepSeek models indicates a proactive approach to addressing user feedback and improving model performance [3]

Core Insights - DeepSeek Company released its new model architecture DeepSeek-V3.2, which has garnered significant industry attention [1] - Cambricon announced its adaptation to DeepSeek-V3.2 and open-sourced the large model inference engine vLLM [1][2] - The DeepSeek-V3.2-Exp model introduces DeepSeek Sparse Attention, optimizing training and inference efficiency for long texts [1] Company Developments - DeepSeek-V3.2-Exp is an experimental version built on V3.1-Terminus, focusing on sparse attention mechanisms [1] - The official DeepSeek applications and APIs have been updated to V3.2-Exp, with significant price reductions to encourage user testing and feedback [1] - Cambricon's adaptation to DeepSeek-V3.2-Exp indicates prior collaboration and technical communication between the two companies [2] Industry Trends - The rapid adaptation of Cambricon to the new model reflects a significant signal of deep collaboration among top Chinese tech companies [2] - The large model has a substantial size of 671GB, requiring approximately 8-10 hours to download under ideal bandwidth conditions [2] - The collaboration between leading companies in the AI chip and model sectors is seen as a strong example of innovation and cooperation in China's tech industry [2]

Core Viewpoint - The release of DeepSeek-V3.2 by DeepSeek Company and its adaptation by Cambricon signifies a strong collaboration among leading tech firms in China's AI industry, aiming to enhance efficiency in long-text training and inference [2][3][4]. Group 1: Model Release and Features - DeepSeek Company launched the experimental version DeepSeek-V3.2-Exp, which introduces a sparse attention mechanism for optimizing long text training and inference [2]. - The new model has a substantial size of 671GB, requiring approximately 8-10 hours for download under ideal bandwidth conditions [3]. Group 2: Collaboration and Industry Impact - Cambricon's quick adaptation to DeepSeek-V3.2-Exp indicates prior collaboration and communication between the two companies, reflecting a trend of low-profile yet effective partnerships in the tech industry [3]. - The collaboration between leading companies in the AI model and chip sectors is expected to significantly reduce training and inference costs for users, facilitating the emergence of AI applications [4].

Core Viewpoint - The article discusses the current state of the A-share market, highlighting the significant interest in the AI industry and contrasting it with the performance of traditional value stocks, particularly in the consumer sector [5][8][16]. Group 1: Market Trends - Since the macro policy shift a year ago, the A-share market has become a standout performer globally, with major indices like the ChiNext and STAR Market leading the way [5]. - The AI sector has emerged as a hot topic, with investor-created concepts like "Yizhongtian" and "Jilianhai" gaining attention, representing different segments of the AI computing industry [8][9]. Group 2: Stock Performance - Stocks related to the "Yizhongtian" concept, such as New Yisheng and Tianfu Communication, have benefited from global AI infrastructure demands, while "Jilianhai" stocks like Cambricon and Industrial Fulian focus on domestic AI chip and server production [9]. - Notable stock price increases have been observed, with Cambricon briefly surpassing Kweichow Moutai to become the "king of A-shares," and Industrial Fulian's market value exceeding one trillion [9]. Group 3: Investment Sentiment - There is a strong concentration of funds in AI-related stocks, leading to a "stronger getting stronger" effect, while traditional value stocks in sectors like liquor and dairy are experiencing significant underperformance [9][12]. - The article critiques the perception of value investors, referred to as "old Deng," who are often dismissed for their focus on traditional sectors despite their long-term investment logic [10][12]. Group 4: Valuation Concerns - The article raises concerns about the high valuations in the AI sector, with the average P/E ratio for the STAR Market at 67.16 and the ChiNext at 43.40, indicating a significant divergence from fundamental analysis [17]. - The liquidity in the market is noted to be exceptionally high, with daily trading volumes exceeding 30 trillion and margin financing balances over 2 trillion, contributing to the inflation of stock prices [16][17]. Group 5: Long-term Investment Strategy - The author emphasizes the importance of investing in fundamentally strong companies, particularly in the consumer sector, which are currently undervalued, suggesting that long-term holding can yield substantial returns [27]. - The article also highlights successful investments in global tech companies, such as Alibaba and BYD, as examples of how strategic investments can lead to significant gains over time [24][26].

9月29日，寒武纪大宗交易成交1.52万股，成交额2011.72万元，占当日总成交额的0.17%，成交价1323.5 元，较市场收盘价1323.5元持平。 | 交易日期 | | 证券代码 ----- | 成交价(元) 成交金额[万元) 成交量( *) 买入营业部 | | 卖出营业部 | 是否为专场 | | --- | --- | --- | --- | --- | --- | --- | | 025-09-29 | 寒武纪 | 688256 | 1323.5 2011.72 | 52 有限公通运营股份 | 去筒年塑膠纹菌眼 | KO | ...

Core Insights - The main point of the news is the net inflow of capital in the Shanghai and Shenzhen stock markets, amounting to 9.527 billion yuan, with a notable outflow in the Sci-Tech Innovation Board [1] Group 1: Market Overview - The total net inflow of capital in the Shanghai and Shenzhen markets was 9.527 billion yuan, with the Sci-Tech Innovation Board experiencing a net outflow of 4.624 billion yuan [1] - A total of 210 stocks saw net inflows, while 377 stocks experienced net outflows [1] Group 2: Stock Performance - On the Sci-Tech Innovation Board, 404 stocks increased in value, with three stocks hitting the daily limit up, including Pinming Technology and Fuzhi Environmental Protection [1] - Among the stocks with net inflows, 9 stocks had inflows exceeding 100 million yuan, with Lanqi Technology leading at 302 million yuan [2] - The stocks with the highest net outflows included Haiguang Information, which saw a net outflow of 1.077 billion yuan, followed by Chip Origin and Crystal Integration with outflows of 297 million yuan and 267 million yuan, respectively [1][2] Group 3: Continuous Capital Flow - There were 39 stocks with continuous net inflows for more than three trading days, with Hanwujing leading at 29 consecutive days [2] - A total of 166 stocks experienced continuous net outflows, with Aibo Medical leading at 15 consecutive days [2] Group 4: Detailed Stock Data - The top stocks by net inflow included: - Lanqi Technology: 301.69 million yuan, 4.29% inflow rate, 7.93% increase [2] - Zhongwei Company: 271.03 million yuan, 4.45% inflow rate, 5.17% increase [2] - Wealth Trend: 155.91 million yuan, 9.66% inflow rate, 7.51% increase [2] - The stocks with the highest net outflows included: - Haiguang Information: 1.077 billion yuan, 1.26% decrease [1] - Chip Origin: 297 million yuan, 2.67% decrease [1] - Crystal Integration: 267 million yuan, 2.67% decrease [1]

Core Insights - The average stock price of A-shares is 13.62 yuan, with 168 stocks priced over 100 yuan, an increase of 3 stocks from the previous trading day [1] - The Shanghai Composite Index closed at 3862.53 points, up 0.90%, while stocks over 100 yuan had an average increase of 2.14%, outperforming the index by 1.24 percentage points [1][2] - The top three stocks with the highest closing prices over 100 yuan are Kweichow Moutai at 1460.86 yuan, Cambricon at 1323.50 yuan, and Gigabyte at 588.20 yuan [1][2] Market Performance - In the past month, stocks priced over 100 yuan have averaged a 15.58% increase, compared to a 0.49% increase in the Shanghai Composite Index [2] - The best-performing stocks in this category include Haibo Technology, Tianpu Co., and Demingli, with increases of 173.99%, 159.39%, and 89.88% respectively [2] - Year-to-date, the average increase for these high-priced stocks is 114.09%, significantly outperforming the Shanghai Composite Index's 98.85% [2] Stock Distribution - Among the stocks priced over 100 yuan, 42 are from the main board, 48 from the ChiNext board, 3 from the Beijing Stock Exchange, and 75 from the Sci-Tech Innovation Board, with the latter accounting for 44.64% of the total [3] - The electronic industry has the highest representation with 62 stocks, making up 36.90% of the total, followed by the computer industry with 19 stocks (11.31%) and the pharmaceutical and biological industry with 17 stocks (10.12%) [2][3] Institutional Ratings - Four stocks priced over 100 yuan received buy ratings from institutions today, including WuXi AppTec, Wuliangye, and Mindray Medical, with WuXi AppTec being newly covered by institutions [3]

以最新收盘价计算，科创板平均股价为42.37元，其中股价超100元的有75只，股价最高的是寒武纪-U。 证券时报·数据宝统计显示，科创板股今日上涨的有404只，下跌的有177只，以收盘价为基准测算，科 创板平均股价为42.37元，其中，收盘价超过100元的有75只，股价在50元至100元之间的有147只，股价 在30元至50元的有145只。 科创板股中，收盘价最高的是寒武纪-U，今日报收1323.50元，下跌1.16%，其次是茂莱光学、源杰科技 等，最新收盘价分别为424.80元、410.23元。 科创板百元股中，今日平均上涨2.31%，具体来看，今日上涨的有51只，涨幅居前的有海博思创、源杰 科技、开普云等。下跌的有23只，跌幅居前的有合合信息、百利天恒、腾景科技等。 今日有1只股收盘价首次突破百元大关，京仪装备最新收盘价107.49元，今日上涨9.18%，全天换手率 7.03%，成交额8.59亿元，全天主力资金净流出1716.43万元。 资金流向方面，科创板百元股今日主力资金合计净流出11.21亿元，净流入资金居前的有澜起科技、中 微公司、财富趋势等，净流入资金分别为30168.89万元、27102.90 ...

Core Viewpoint - The article discusses the remarkable rise of the innovative drug company,药捷安康, which has seen its stock price soar despite having no revenue from product sales, primarily due to speculation and market excitement surrounding its lead product, Tinengotinib [1][12][20]. Company Overview - 药捷安康 is focused on developing small molecule therapies for cancer, inflammation, and cardiovascular diseases, founded in 2014 by 吴永谦, who has extensive experience in the biopharmaceutical industry [3][4]. - The company’s lead product, Tinengotinib, is a multi-target kinase inhibitor undergoing clinical trials for cholangiocarcinoma (CCA) and metastatic castration-resistant prostate cancer (mCRPC) [4][5]. Financial Performance - 药捷安康 reported a revenue of 40.55 million RMB in 2023, with the majority coming from government subsidies and bank interest, resulting in a minimal actual revenue from operations [12][13]. - The company has been operating at a loss, with losses of 343 million RMB in 2023 and projected losses of 275 million RMB in 2024 [17]. Market Potential - The global market for CCA drugs is projected to reach 2 billion USD in 2024, while the prostate cancer drug market is expected to grow to 19.3 billion USD [10]. - 药捷安康's Tinengotinib is the first and only drug in clinical registration for treating FGFR-inhibitor resistant CCA patients and shows potential for treating other cancers [5][10]. Stock Performance - 药捷安康's stock price surged to 679.5 HKD per share, a 50-fold increase from its initial offering price of 13.15 HKD, driven by market speculation and positive clinical trial news [2][18]. - The stock's rapid rise is attributed to limited circulating shares, with only about 1.38% of total shares available for trading, making it susceptible to price manipulation [21][22]. Clinical Development - The company is advancing Tinengotinib through clinical trials, with plans to recruit a marketing director to lead commercialization efforts as it aims for market entry in China and potentially in the US and EU [17][20]. - The drug has received clinical trial approval for use in combination with another treatment for breast cancer, which has further fueled investor interest [20]. Research and Development - 药捷安康 has reduced its R&D expenses significantly, with a focus on Tinengotinib, which accounted for a large portion of its R&D budget [15][22]. - The company is also developing additional candidates, including treatments for various cancers and inflammatory diseases, although these are still in early stages [10][12].