CHINA STOCK
ALL Announcement IPOs Research Market Company

China Galaxy Securities: DeepSeek innovation brings new opportunities for automotive intelligence Chongqing Sokon Industry Group Stock (601127.SH) and other vehicle recommendations.

China Galaxy Securities released a research report stating that DeepSeek has significantly improved training efficiency and reduced computing costs under limited hardware conditions, while achieving a significant improvement in model performance. In terms of intelligent driving applications, the demand for on-vehicle computing power has decreased, leading to a reduction in cloud-based model training costs and significant cost savings. In addition, Deepseek is an open-source model that reduces technical barriers, benefiting small and medium-sized manufacturers or manufacturers just starting to focus on intelligent driving. In terms of whole vehicles, BYD Company Limited (002594.SZ, 01211), Chongqing Sokon Industry Group Stock (601127.SH), Ideal Cars (02015), and Chongqing Changan Automobile (000625.SZ) are recommended. Beneficiaries include GEELY AUTO (00175), XPENG-W (09868), LEAPMOTOR (09863), and XIAOMI-W (01810). In terms of components, Huizhou Desay SV Automotive (002920.SZ), Bethel Automotive Safety Systems (603596.SH), Keboda Technology (603786.SH), ROBOSENSE (02498), and Beijing Jingwei Hirain Technologies (688326.SH) are recommended. Beneficiaries in this sector include BYD ELECTRONIC (00285), Longhorn Auto (301488.SZ), and HORIZONROBOT-W (09660). The main points of China Galaxy Securities are as follows: Deepseek has significantly improved training efficiency and reduced computing costs under limited hardware conditions, while achieving a significant improvement in model performance. In terms of intelligent driving applications, the decrease in on-vehicle computing power demands has led to a reduction in cloud-based model training costs, resulting in significant cost savings. In addition, Deepseek is an open-source model that reduces technical barriers, benefiting small and medium-sized manufacturers or manufacturers just starting to focus on intelligent driving. On January 20, 2025, DeepSeek officially released the R1 model and simultaneously open-sourced the model weights. In third-party benchmark tests, the R1 continued the excellent performance of the previous DeepSeek basic language model V3, performing on par with the top global large model OpenAIol. DeepSeek-R1 has achieved a leading performance of large models and a substantial optimization of costs through the innovation of its algorithm architecture, bringing new insights to the development and application of automotive intelligence technology represented by intelligent driving and intelligent cabins. It is expected to promote the maturity of intelligent technology and reduce research and development costs, accelerating the transformation of the automotive industry to intelligence. Intelligent driving Bringing technological inspiration, accelerating the "intelligent driving equalization": DeepSeek-R1 has innovatively improved training efficiency and reduced training costs through algorithms such as MLA, DeepSeekMoE, load balancing without auxiliary losses, and MTP. By applying the innovative algorithm architecture of DeepSeek-R1 in the development of intelligent driving systems, it is expected to improve the training efficiency and reduce the costs of intelligent driving systems, thereby accelerating the progress of technology for domestic brands with currently high market share in the new energy market but temporarily lagging in intelligent technology. It can drive the mass adoption of advanced intelligent driving functions and push the prices of high-end intelligent driving function models downward, promoting the trend of "intelligent driving equalization." Due to the higher requirements for safety, stability, and latency in intelligent driving systems compared to language models, leading intelligent driving companies can still establish a strong technological moat based on their advantages in technology talent reserves, high-quality driving data accumulation, system maturity, and technological research and development. Therefore, the impact of DeepSeek on the competitive landscape of the intelligent driving industry is considered to be small. Its significance lies in lowering the threshold for some advanced intelligent driving technologies and funds to accelerate the arrival of the "intelligent driving equalization" era. In addition, the training mode of DeepSeek's cold start data + unsupervised reinforcement learning provides a new idea for the training of intelligent driving systems. If applied, it is expected to reduce annotation costs and improve model generalization ability. By constructing a world model with DeepSeek's multimodal capabilities for simulating long-tail training scenarios and training, it is expected to promote the maturity of intelligent driving system technology. Intelligent cabin Automotive companies are gradually adopting it, and the intelligent cabin experience is expected to improve: DeepSeek-R1's core advantages in interactive experience, open source, and cost are well suited to the requirements of automotive intelligent cabins. Based on the interoperability of large language models, it can also be directly migrated to intelligent cabin voice systems, making it highly attractive to automotive companies. On February 6th, Geely officially announced that its self-developed Xingrui large model and DeepSeek-R1 large model have been deeply integrated, becoming the first automotive company to implement DeepSeek. Subsequently, companies such as Jike, Dongfeng, Zerun, Great Wall, Changan, and BYD Company Limited have successively adopted DeepSeek for intelligent cabins, with DeepSeek expected to further enhance the intelligent cabin experience in the automotive industry.
21/02/2025

Dongxing: Crude oil prices fell in February, and the supply of oil products in the United States decreased.

Dongxing released a research report stating that as of February 14th, the crude oil prices had fallen compared to the previous month. In January, the monthly average prices of OPEC and Chinese crude oil spot prices both increased month-on-month. As of February 7th, the operating capacity utilization rate of U.S. refineries continued to decline, and the supply of petroleum products decreased. In January, OPEC production decreased month-on-month, while U.S. crude oil exports increased compared to the previous month. In December, Chinese crude oil exports saw a significant increase. The main points of Dongxing's analysis are as follows: Brent crude oil prices fell month-on-month, ESPO crude oil spot prices also fell, while Chinese crude oil spot prices increased month-on-month. As of February 14th, Brent and WTI crude oil futures settlement prices were $74.74 per barrel and $70.74 per barrel respectively, down 6.48% and 8.72% month-on-month; WTI and Brent spot prices were $71.29 per barrel and $75.02 per barrel, down 9.55% and 7.39% respectively from the previous month; ESPO spot price was $69.05 per barrel, down 7.19% month-on-month. In January, OPEC spot price was $79.38 per barrel, up 8.63% month-on-month; Chinese spot price (South Sea) was $70.24 per barrel, up 9.09% month-on-month, and Chinese spot price (Victory) was $76.92 per barrel, up 5.68% month-on-month. OPEC production decreased month-on-month, U.S. refinery operating capacity utilization rate continued to decline, and the supply of petroleum products decreased. In January, OPEC crude oil production reached 26,678 thousand barrels per day, down 122 thousand barrels per day month-on-month, a decrease of 0.46%, and an increase of 296 thousand barrels per day year-on-year, an increase of 1.12%. As of February 7th, the average weekly operating capacity utilization rate of U.S. refineries was 85.00%, down 6.70 percentage points month-on-month; the supply of petroleum products was 19,624 thousand barrels per day, down 1,049 thousand barrels per day month-on-month, a decrease of 5.07%. U.S. crude oil exports increased month-on-month, and Chinese crude oil exports saw a significant increase. In January, the average daily export quantity of U.S. crude oil was 3,937.60 thousand barrels, up 45.10 thousand barrels month-on-month, an increase of 1.16%; in December, Chinese crude oil exports were 417,225.08 tons, up 65.81% month-on-month. Risk factors: Geopolitical risks; risks of significant fluctuations in energy prices; risks of demand falling short of expectations.
21/02/2025

Orient: AI Empowering Biopharmaceuticals, Technology and Policy Support

Orient released a research report stating that since the Spring Festival, Deepseek has sparked continuous attention to the AI industry chain in the market. ARK's "Big Ideas 2025" report points out that AI and multi-omics will trigger a revolution in drug development, molecular diagnostics, and disease treatment, leading to a continuous increase in the popularity of AI healthcare. Since 2024, there has been a dense release of AI-related policies that comprehensively support the development of AI healthcare from the perspectives of development, application scenarios, and standard setting. In terms of market size, the AI healthcare industry in China was approximately 8.8 billion yuan in 2023, and is expected to grow to 315.7 billion yuan by 2033, with a high CAGR of 43.1%. AI Applications: Drug discovery, medical imaging, and healthcare information technology AI can integrate with various aspects of the healthcare field, with drug discovery, medical imaging, and healthcare information technology showing great potential and being the focus of future AI healthcare development. 1) AI pharmaceuticals: Currently focused on the early stages of drug discovery, AI pharmaceuticals are expected to solve the challenges of high costs, long cycles, and low success rates faced by traditional drug development. However, the scarcity of high-quality data still hinders the "Deepseek" moment in the biopharmaceutical industry. 2) Medical imaging: By using artificial intelligence technology to analyze and interpret medical images, AI can improve the efficiency of radiologists, reduce misdiagnosis rates, and increase accuracy. In recent years, the application of AI medical imaging has been rapidly expanding, with increasing market demand, focusing mainly on cardiovascular and pulmonary diseases, with domestic companies leading the market. 3) Healthcare information technology: Using AI and other technologies to establish an intelligent information management system, healthcare information technology can enhance hospital management efficiency, diagnosis and treatment efficiency, and patient healthcare experience. Intelligent healthcare for medical staff development has been matured under the catalysis of relevant policies. With the introduction of new standards for intelligent healthcare evaluation, the implementation of AI healthcare information technology is expected to accelerate. As AI technology iterates rapidly and with stimulation from the domestic policy environment, there is a rapid release of demands in the healthcare industry. Currently, AI can be combined with various aspects of the healthcare industry to greatly improve efficiency, reduce costs, and better allocate R&D and medical resources. From the application perspective, AI pharmaceuticals, medical imaging, and healthcare information technology have shown great potential. It is recommended to pay attention to companies in related sectors that have a domestic layout: 1) AI Pharmaceuticals: WuXi AppTec (02359), Pharmaron Beijing (03759), Shanghai Medicilon Inc. (688202.SH), PharmaResources (301230.SZ), Hitgen Inc. (688222.SH), PharmaBlock Sciences (300725.SZ), XTALPI-P (02228), etc. 2) Medical Imaging: Shanghai United Imaging Healthcare (688271.SH), Guangzhou LBP Medicine Science & Technology (688393.SH), Shanghai Labway Clinical Laboratory (301060.SZ), Lepu Medical Technology (300003.SZ), XUNFEIHEALTH (02506), Chison Medical Technologies (688358.SH), Beijing Wandong Medical Technology (600055.SH), etc. 3) Healthcare Information Technology: Winning Health Technology Group (300253.SZ), Goodwill E-Health Info (688246.SH), B-Soft Co., Ltd. (300451.SZ), YIDU TECH (02158), Meinian Onehealth Healthcare Holdings (002044.SZ), etc. Risk Warning: Risks associated with technological iteration and policy changes.
21/02/2025

Soochow: Figure releases VLA large-scale model Helix, significantly reducing the need for training data.

Soochow released a research report stating that after ending its cooperation with OpenAI in mid-February, local time on February 20, 2025, Figure announced the launch of its self-developed general embodied intelligent model Helix and demonstrated the ability of two Siasun Robot & Automation to work together. The report suggests paying attention to investment opportunities in the Figure chain; additionally, it highlights the importance of the coordination between the ventral large model (VLM) and the small brain in direct reuse. Soochow's main points are as follows: Helix's large model has multiple firsts in the industry and excellent performance prospects for commercialization Helix is a general visual-language-action (VLA) model that combines perception, language understanding, and control abilities. Innovative achievements by Helix include: Full-body control - the first VLA model to control the upper extremities of a humanoid Siasun Robot & Automation at high speed, including the head, trunk, and hands/fingers; Multiple Siasun Robot & Automation collaboration - Helix is the first VLA model that can simultaneously control two Siasun Robot & Automation, enabling them to collaborate to solve a long-term operation task; Excellent grasping ability - capable of picking up any small item, even those that have never been seen or touched before; Single neural network - Helix uses a single set of neural network weights to learn all behaviors without any specific task fine-tuning; Rapid commercialization capability - Helix is the first VLA model that can run fully on embedded low-power GPUs, giving it excellent commercialization potential. Pioneering a dual-system collaboration architecture to achieve coordinated "fast thinking" and "slow thinking" Figure stated that the problem faced by the previous large models of humanoid Siasun Robot & Automation was that the VLM backbone had strong generality but slow speed, while the visual motion strategy of Siasun Robot & Automation was fast but lacked generality. Helix innovatively adopts a decoupling architecture of "system 1" + "system 2," where the two systems can communicate with each other after end-to-end training. System 1 (S1) is a fast reactive visual motion strategy that can convert the potential semantic representation generated by S2 into precise continuous Siasun Robot & Automation actions at a frequency of 200 Hz, and System 2 (S2) is a VLM pre-trained with internet data, running at a frequency of 7-9 Hz for scene understanding and language understanding, achieving generalization in different objects and situations. This architecture brings many highlights to Helix: High-speed and generalization capability - equivalent speed to single-task cloning and zero-shot generalization; Direct output of continuous control over high-dimensional action space, avoiding the use of complex action labeling schemes; Simple architecture of S1+S2; Focus separation: Decoupling of S1 and S2 allows for separate iteration of systems. Helix achieved strong generalization capabilities efficiently with only 500 hours of training Figure stated that Helix was trained using approximately 500 hours of high-quality supervised data, only a small part of the size of the previously collected VLA dataset, and did not rely on data collection from multiple Siasun Robot & Automation entities or multiple training phases. The core advantage of Figure this time lies in solving the current humanoid Siasun Robot & Automation generalization bottleneck - insufficient training data. In the past, the amount of data collected through remote control and virtual simulation was severely insufficient, costly, and difficult to address the generalization problem. This time, through model innovation, Figure significantly reduced the training data requirements for Siasun Robot & Automation, potentially accelerating the overall development of the industry. Risk warnings: the progress of mass production of humanoid Siasun Robot & Automation is slower than expected, the technological iteration of humanoid Siasun Robot & Automation is slower than expected, and macroeconomic risks.
21/02/2025

Guotai Junan: Since 2025, the oil shipping industry has experienced a rebound in market conditions and is expected to provide unexpected and counter-cyclical opportunities in the future.

Guotai Junan released a research report stating that it maintains a "buy" rating on the oil shipping industry and is paying attention to counter-cyclical opportunities. In the next two years, the irregular supply of ships will remain rigid. Among them, oil shipping will benefit from the global increase in crude oil production, with supply and demand expected to be better than market expectations, and also having the option of falling oil prices. Since 2025, the sanctions on shadow fleets have tightened, contributing to the recovery of the oil shipping industry. The future geopolitical situation may evolve or diverge, providing opportunities for unexpected demand and counter-cyclical opportunities. Guotai Junan's main points are as follows: Regular shipping (container shipping): Gradually returning to a new normal, focusing on the power of alliances The container shipping industry has been dominated by two cycles of high prosperity in the past five years, driven by events. The profit center of container shipping companies has significantly increased, achieving high dividends and low valuations that lead to high dividend yields. The future trend of container shipping prosperity depends on exports and capacity digestion. The role of alliances in capacity digestion is crucial, and future alliance restructuring may affect capacity digestion capabilities. Irregular shipping (oil shipping & dry bulk): Continued rigid supply, expecting unexpected demand Crude oil shipping: Tightening sanctions on shadow fleets have driven the recovery of the oil shipping industry. It is reiterated that increased crude oil production will benefit oil shipping, with future supply and demand expected to be better than expected, and also having the option of falling oil prices. Refined oil shipping: The continuous relocation of global refineries to the east will continue to drive demand for refined oil shipping, potentially offsetting supply delivery pressure, and the prosperity is expected to continue or exceed market expectations. Dry bulk shipping: Rigorous supply is expected to continue in the coming years, with expectations for increased mining production and changes in trade structure, leading to a gradual recovery in prosperity. Geopolitical situations continue to impact the global shipping industry In the past three years, geopolitical situations have continued to impact the shipping industry, driving changes in trade structure and unexpected shipping demand. In the second half of 2024, this also led to the oil shipping industry undergoing pressure testing. Since 2025, sanctions on shadow fleets have tightened, contributing to the recovery of the oil shipping industry. The future geopolitical situation may evolve or diverge, providing opportunities for unexpected demand and counter-cyclical opportunities. Risk warnings: Risks of economic fluctuations, geopolitical risks, tariff policies and trade sanctions risks, oil price risks, risks of environmental policy enforcement falling short of expectations, safety accident risks, etc.
21/02/2025

Cinda: Chinese New Year's off-season affect January parcel growth. SF Express (06936) leads in parcel growth speed.

Cinda released a research report stating that the off-peak period during the Chinese New Year in January affected the growth rate of the company's express delivery volume. S.F. Holding had a leading growth rate in business volume, with YTO Express, Yunda, STO Express, and SF Express achieving 2.268 billion, 2.013 billion, 2.023 billion, and 1.33 billion express business volumes respectively in January. In terms of growth rate, SF Express was at 16.0% > STO Express 11.8%> YTO Express 5.5%> Yunda 2.9%. In terms of business revenue, YTO Express had revenue of 5.34 billion (+1.5%), Yunda 4.069 billion (-8.4%), STO Express 4.169 billion (+5.1%), and SF Express 20.763 billion (+6.5%) in January. Regarding competition, in January, the e-commerce express delivery prices saw a slight increase, with SF Express's price increasing by 7.3%. Cinda's main points are as follows: Volume situation: During the Spring Festival, express delivery volume maintained high growth, with SF Express leading in business volume growth in January. 1) Upstream driving and industry: Since 2025, the volume of express deliveries has maintained a high growth trend. According to data from the State Post Bureau, since the beginning of the Spring Festival (January 14th to February 18th), the nationwide postal express industry received 15.549 billion express packages, an increase of 29% compared to the same period in 2024; and delivered 15.969 billion express packages, an increase of 31.1% compared to the same period in 2024. 2) Company's business volume: The off-peak period during the Chinese New Year in January affected the growth rate of the company's express delivery volume, with S.F. Holding leading in business volume growth. In January, YTO Express, Yunda, STO Express, and SF Express achieved 2.268 billion, 2.013 billion, 2.023 billion, and 1.33 billion express business volumes respectively. In terms of growth rate, SF Express was at 16.0%> STO Express 11.8%> YTO Express 5.5%> Yunda 2.9%, with S.F. Holding leading in business volume growth in January. Price situation: In January, the industry's competitive order remained stable, with prices showing a slight increase. 1) Express business income: In January, SF Express's logistic business income increased by 6.5% year-on-year. In terms of express business income, YTO Express had revenue of 5.34 billion (+1.5%), Yunda 4.069 billion (-8.4%), STO Express 4.169 billion (+5.1%), and SF Express 20.763 billion (+6.5%). 2) Unit price: The unit price in January showed a slight increase. YTO Express was at 2.35 yuan (down 3.8% year-on-year, up 2.6% month-on-month); Yunda was at 2.02 yuan (down 11.0% year-on-year, down 0.5% month-on-month); STO Express was at 2.06 yuan (down 5.9% year-on-year, up 2.0% month-on-month); and SF Express was at 15.61 yuan (down 8.2% year-on-year, up 7.3% month-on-month). Core summary and outlook: The industry's single volume still has growth potential. Attention to changes in the competitive landscape in 2025. 1) Business volume: The scale expansion growth rate of e-commerce express deliveries is still ongoing. Against the backdrop of the further rise of live e-commerce, on one hand, the penetration rate of online shopping consumption is further increasing, and on the other hand, the sinking and fragmentation of online shopping consumption behaviors are driving down the online shopping value of single express package physical goods. The express delivery industry still has excess growth relative to upstream e-commerce. It is expected that the express delivery industry will achieve a growth rate of 10 to 15% in 2025. 2) Competition order: In January, the e-commerce express delivery prices saw a slight increase, with SF Express's price rising by 7.3%. According to ZTO Express's 2024 third-quarter performance announcement, the company plans to regain market share, expand its leading advantage in business volume, and achieve a reasonable level of profitability. In the context of increased demands for market share by leading companies, there may be a return to price competition in 2025. The industry landscape changes in the background of price competition are to be observed. Investment recommendation 1) Regarding the franchising system, valuation may have been adjusted, and attention should be given to changes in the industry landscape in the background of price competition. Recommended investment in ZTO Express (02057), YUNDA Holding Group (002120.SZ), YTO Express Group (600233.SH), and attention to STO Express Co., Ltd.(002468.SZ). 2) Regarding the direct-sales system, it is recommended to invest in S.F. Holding (002352.SZ,06936). As a comprehensive leader in express delivery logistics, the company's operational and cash flow inflection points are expected to usher in a new development stage. In the short term, with the advancement of the company's network integration, steady profit margin improvement may drive relatively high performance growth. In the medium to long term, international business is expected to open up a new growth curve. Risk factors Demand for physical goods online shopping is lower than expected; intensified competition in e-commerce express delivery prices; decreased stability of end-line franchisees.
21/02/2025

China Galaxy Securities: Policy measures strengthen control, supply and demand dynamics improve rare earths bottoming out rebound

China Galaxy Securities released a research report stating that the rare earth policy is further tightening supply control, AI empowering humanoid Siasun Robot & Automation is accelerating its implementation, and low-altitude economy is opening up long-term space for rare earth demand. The bank calculated that by 2024-2025, as domestic indicators slow down and supply-side control tightens, the pressure of rare earth oversupply will gradually ease, reaching supply-demand balance by 2026. This will drive the currently bottomed rare earth prices to rise, helping to significantly improve the performance of the rare earth magnetic materials industry. Furthermore, the strengthening of rare earth control policies will further enhance the leading position of rare earth groups in the industry, favoring the increase in their valuation. Event: On February 19, in order to implement the "Rare Earth Management Regulations," the Ministry of Industry and Information Technology released the "Rare Earth Mining and Rare Earth Smelting Separation Total Control Management Measures (Temporary) (Draft for Public Comments)" (referred to as the "Total Control Consultation Draft") and "Rare Earth Product Information Traceability Management Measures (Temporary) (Draft for Public Comments)" (referred to as the "Product Traceability Consultation Draft"), soliciting public opinions. The main points of China Galaxy Securities are as follows: The consultation draft is a refinement and implementation of the "Rare Earth Management Regulations." In June 2024, the State Council issued the "Rare Earth Management Regulations," which intensified management in areas such as rare earth resource protection, rare earth product information traceability, rare earth import and export, and penalties for violations, further restricting "illegal rare earths" and expanding the scope of rare earth product management. The consultation draft released by the Ministry of Industry and Information Technology further elaborates on the relevant provisions of the "Regulations" in terms of rare earth total control and product information traceability, effectively protecting and reasonably developing rare earth resources, standardizing rare earth mining, smelting separation production activities, strengthening rare earth product information traceability management, and promoting the high-quality development of the rare earth industry. Imported ores and monazite are included in the separation and smelting quota management scope. The "Total Control Consultation Draft" defines the sources of rare earths in the smelting separation as domestically mined rare earth ore products, imported rare earth ore products, monazite concentrates, and other rare earth ore products obtained as by-products of other rare earth-containing minerals. Previously, imported rare earth ores, monazite, etc. were not included in the rare earth mining and smelting separation quotas. The consultation draft now includes imported rare earth ores and monazite in the smelting separation quota management, further enhancing the supervision of rare earth products. According to data from the General Administration of Customs, China imported 133,000 tons of rare earths in 2024, with monazite accounting for about 25,000 tons of REO; the global rare earth production in 2024 is approximately 466,000 tons of REO, with imports and monazite contributing 34% to the rare earth supply. Compliant production entities are clearly defined as large rare earth groups. The "Total Control Consultation Draft" clearly defines the scope of compliant rare earth mining and smelting separation entities and stipulates that the Ministry of Industry and Information Technology, in conjunction with relevant departments, will determine and publicize rare earth mining and smelting separation enterprises, which should be large rare earth groups and their subsidiaries established by the state; other organizations or individuals are not allowed to obtain rare earth quotas or engage in related production activities. This requirement will place restrictions on smelting separation capacity outside of large rare earth groups and effectively curb the disorderly mining and waste of resources. Product traceability effectively enhances the supervision of the entire distribution chain of rare earths. The applicable subjects of the "Product Traceability Consultation Draft" include enterprises engaged in rare earth mining, smelting separation, metal smelting, comprehensive utilization, and export of rare earth products in China, with the product range including rare earth ore products, various rare earth compounds, various rare earth metals and alloys, etc. The draft requires rare earth enterprises to enter data into the rare earth traceability system by the 10th of each month; encourages rare earth enterprises to establish internal rare earth traceability systems; and promotes the use of traceability codes by rare earth enterprises, applying for product traceability codes through the rare earth traceability system to label each unit of product for traceability. Risk warning: 1) Risks of domestic economic recovery falling short of expectations; 2) Risks of downstream demand falling short of expectations; 3) Risks of overseas mine capacity being put into operation faster than expected; 4) Risks of a significant drop in rare earth prices.
21/02/2025

Minsheng Securities: Quantum computing may become a disruptive force and bottleneck for and is expected to accelerate the application of the AI industry.

Minsheng Securities released a research report stating that quantum computing is expected to overturn classical computing architecture and become a disruptive force in solving the AI computing bottleneck, or become an important lever for developing new quality productivity. Sorting out the industry chain related targets, including Quantumctek Co., Ltd. (688027.SH) - quantum computer, Guangzhou Hexin Instrument (688622.SH) - dilution refrigerator and other quantum computing related targets, as well as CETC Cyberspace Security Technology (002268.SZ), Jilin University Zhengyuan Information Technologies (003029.SZ), Koal Software (603232.SH), C*Core Technology (688262. SH), Zhejiang Orient Financial Holdings Group (600120.SH), Hengtong Optic-Electric (600487.SH) and other quantum encryption communication targets. The main points of Minsheng Securities are as follows: Microsoft released its first quantum computing chip. According to Tencent Technology reports, on February 20, after nearly 20 years of research, Microsoft launched its first quantum computing chip Majorana1 in the United States on Wednesday. Microsoft stated that developing Majorana1 required the creation of a new form of material state, known as "topological body". Majorana1 is as small as 0.01 millimeters wide and has achieved the placement of 8 topological qubits on a single chip. In the future, this chip will be able to expand to millions of qubits. This chip could perform tasks that all the current computers in the world working together cannot achieve with a 100,000 qubit quantum computer. Using a new architecture to produce more reliable computing units. The core of a quantum computer is quantum bits (qubits), which are the information units in quantum computing, similar to the binary system used by computers. However, qubits are fragile and very sensitive to environmental noise, which can lead to computing errors or data loss - a destructive outcome for computers. This is the core contradiction that has slowed down the development of quantum computing. In order to solve this problem, Microsoft's quantum chip has adopted a new solution by creating the so-called "world's first topological body" to observe and control Majorana particles, thereby generating more reliable and scalable qubits. Microsoft solves architecture issues through materials. Microsoft stated that this topological body uses indium arsenide (a semiconductor) and aluminum (a superconductor) to design and build the topological conductor wires atom by atom - the "quantum transistor of the quantum era". Microsoft stated in a blog post, "Developing suitable materials for creating exotic particles and their related topological states is extremely difficult, which is why most quantum research focuses on other types of qubits." Quantum chips are expected to accelerate the application of the AI industry. Microsoft CEO Nadella predicts that when AI is combined with quantum computing, quantum computing can be used to generate synthetic data; then, AI can use this data to train better models for application in complex fields such as chemistry and physics. The company believes that when the power of quantum computing is combined with AI, people can create new materials or molecules through language, get answers directly without speculation or experimentation. Risk Warning: Policy implementation falls short of expectations, capital expenditures fall short of expectations, and intensified competitive landscape.
21/02/2025

Shenwan Hongyuan Group: Opening years of the stove are impressive, and the double-line prosperity accelerates after the Spring Festival.

Shenwan Hongyuan Group released a research report stating that according to data from AVCLOUD, offline sales of range hoods from W01-07 (up to February 16) totaled 101,300 units, an increase of 23.7% year-on-year, with sales revenue reaching 398 million yuan, up by 30.0% year-on-year; gas stoves totaled 115,900 units, up by 13.0% year-on-year, with sales revenue of 236 million yuan, an increase of 27.1% year-on-year; in the online market, according to AVCLOUD data, range hoods totaled 899,700 units, an increase of 30.3% year-on-year, with sales revenue of 1.115 billion yuan, up by 16.6% year-on-year; gas stoves totaled 1.275 million units, up by 25.1% year-on-year, with sales revenue of 789 million yuan, an increase of 19.5% year-on-year, showing a good performance in the range hood category. Shenwan Hongyuan Group's main points are as follows: Clearing the haze, the policy of replacing old with new activates the market On July 25, 2024, the National Development and Reform Commission and the Ministry of Finance issued a notice on "Several Measures to Support Large-scale Equipment Renewal and Consumption of Old Products", which mentions the arrangement of around 300 billion yuan of long-term special national debt funds to support large-scale equipment renewal and the replacement of old products. In addition to the 5 subsidized categories of color TV, refrigerator, washing machine, air conditioner, and computer that were already involved in the appliance "replace old with new" policy in 2009, this time three categories, namely water heaters, household stoves, and range hoods, have been added to a total of 8 subsidized categories. According to AVCLOUD's summary data, in 2024, range hoods achieved retail sales of 36.2 billion yuan, an increase of 14.9% year-on-year, with retail volume reaching 20.83 million units, an increase of 10.5% year-on-year; gas stoves achieved retail sales of 20.4 billion yuan, an increase of 15.7% year-on-year, with retail volume of 24.29 million units, an increase of 15.2% year-on-year; electric water heaters achieved retail sales of 23.6 billion yuan, an increase of 1.3% year-on-year, with retail volume of 18.02 million units, an increase of 2.5% year-on-year; gas water heaters achieved retail sales of 29.6 billion yuan, an increase of 8.8% year-on-year, with retail volume of 13.43 million units, an increase of 8.7% year-on-year. The offline retail sales of range hoods, gas stoves, and water heaters in W01-W34 (as of August 26) were -4.0%/-0.6%/+3.9% year-on-year, while offline retail sales were -9.5%/-1.3%/-1.5% year-on-year, indicating a significant boost in new kitchen appliance categories added to subsidies after the policy was implemented. The performance of the range hood category has been good since the beginning of the year, with sales accelerating significantly after the Spring Festival in both online and offline channels. In the offline market, according to AVCLOUD data, in January 2025, range hood sales reached 92,200 units, an increase of 2.7% year-on-year, with sales revenue of 359 million yuan, an increase of 3.6% year-on-year; gas stoves sold 103,400 units, a decrease of 4.4% year-on-year, with sales revenue of 210 million yuan, an increase of 5.3% year-on-year; online range hoods sold 600,400 units, an increase of 14.2% year-on-year, with sales revenue of 739 million yuan, a decrease of 1.5% year-on-year; gas stoves sold 847,000 units, an increase of 12.3% year-on-year, with sales revenue of 526 million yuan, an increase of 3.1% year-on-year. The sales revenue of multi-category markets only achieved single-digit year-on-year growth mainly due to the misalignment of the Spring Festival in January 2025, while sales data significantly accelerated after the Spring Festival. According to AVCLOUD data, range hoods in the offline market from W01-07 (up to February 16) totaled 101,300 units, an increase of 23.7% year-on-year, with sales revenue reaching 398 million yuan, an increase of 30.0% year-on-year; gas stoves totaled 115,900 units, an increase of 13.0% year-on-year, with sales revenue of 236 million yuan, an increase of 27.1% year-on-year; in the online market, according to AVCLOUD data, range hoods totaled 899,700 units, an increase of 30.3% year-on-year, with sales revenue of 1.115 billion yuan, an increase of 16.6% year-on-year; gas stoves totaled 1.275 million units, an increase of 25.1% year-on-year, with sales revenue of 789 million yuan, an increase of 19.5% year-on-year, showing a strong performance in the range hood category. Investment recommendation The policy of replacing old with new has already been implemented in 2025. In addition to continuing the subsidy intensity of the 2024 policy standards, microwave ovens, water purifiers, dishwashers, and rice cookers have been added to the subsidy range. At the same time, individuals who have already received subsidies for certain products in 2024 can continue to enjoy the subsidies when purchasing similar products in 2025. Compared to previous subsidies, this policy subsidy has further broadened the categories and is expected to have a greater appeal to consumers, driving the replacement of old products with new ones. Referring to the time taken for the 2024 policy to take effect, it is expected that the industry's overall sales in the first eight months of this year will still benefit from a low base. In the kitchen appliance sector, Vatti Corporation (002035.SZ) and Hangzhou Robam Appliances (002508.SZ), which have a stable position in traditional categories, as well as Guangdong Vanward New Electric (002543.SZ), which leads in the water heater category, are recommended. Risk warning Risk of fluctuating raw material prices; risk of exchange rate fluctuations; risk of policy fluctuations at home and abroad.
21/02/2025

Open source securities: SAF industry flourishing Chinese companies may benefit from UCO advantage.

Open Source Securities released a research report stating that as the global climate change issue becomes increasingly severe, biofuels and sustainable aviation fuels (SAF) have become key means for the transportation sector to achieve carbon reduction goals due to their outstanding emission reduction characteristics. China's UCO resources advantage, with low cost and high carbon reduction properties, makes SAF products produced in China competitive in the international market. With stable increases in SAF demand in Europe, the UK, and elsewhere, domestic SAF plants may see stable production, thus increasing the profitability of SAF production enterprises in China, benefiting domestic companies with SAF and UCO production capabilities. Key points from the Open Source Securities report are as follows: Bio-diesel industry chain: Bio-diesel, SAF, have outstanding emission reduction characteristics in the transportation sector As the global climate change issue becomes increasingly severe, countries are enacting strict environmental regulations to reduce greenhouse gas emissions, with transportation being a key area for emission reduction. Biofuels, especially bio-diesel and SAF, are seen by various countries as key means to achieve their carbon reduction goals due to their environmental friendliness. Bio-diesel is involved in long-distance shipping, and SAF is involved in air transport, which are sectors that are difficult to electrify. UCO is a type of waste cooking oil that is a major raw material for bio-diesel and SAF production, and it has received attention due to its outstanding emission reduction properties. According to S&P Global data, UCO-produced bio-diesel has the lowest unit carbon emission value, at 19.87gCO2e/MJ. The future of the SAF industry is thriving, and China with UCO resources is expected to become a major supplier of SAF By 2024, global SAF production capacity is limited, but entering 2025, with the EU and UK officially implementing a 2% SAF blending policy, demand for SAF in the global industry will greatly increase. By 2027, CORSIA will require member countries to participate in aviation emissions reduction, and at that time, most countries globally may introduce clear SAF blending ratio policies. By 2050, the International Air Transport Association (IATA) and the Air Transport Action Group (ATAG) have committed to achieving net-zero emissions for the aviation industry, and SAF has the potential to replace a large portion of conventional aviation kerosene. In terms of supply and demand, regions like the EU, UK, and Japan with insufficient raw material supply may import SAF, while regions like China, the US, Indonesia, and Malaysia with ample raw material supply and significant SAF production capacity planning may become major suppliers of SAF. With China's UCO resources advantage of low cost and high carbon reduction properties, SAF products produced in China have strong market competitiveness in the international market. In theory, SAF profitability is strong, and domestic companies with SAF and UCO production capabilities may benefit It is expected that before 2030, the Hydrogenated Esters and Fatty Acids process (HEFA) will be the mainstream production process for SAF. Processes such as Alcohol-to-Jet (ATJ), Fischer-Tropsch (FT), and Power-to-Liquid (PtL) have stronger carbon reduction properties, but their production costs are higher than HEFA, and commercialization is still immature. Due to limited UCO production, the expansion of SAF production capacity using the HEFA process is restricted. PtL employs direct air carbon capture technology, offering the greatest emission reduction potential, and theoretically has no production material bottleneck. If the technology matures and production at scale costs decrease in the future, PtL may become the main SAF production process. According to EASA's calculations, in 2023, the estimated production cost in Europe using the HEFA process for SAF is 1,770 euros/ton, while the average price of SAF is 2,768 euros/ton, with a net profit of nearly 1,000 euros/ton. According to the "Environmental Impact Assessment Report on the 500,000 tons per year bio-based aviation fuel technology renovation and supporting project of Shandong Haike Chemical Engineering Co., Ltd.", Longzhong Information, and calculations by the firm, as of February 7, 2025, the net profit per ton of domestic SAF is approximately 714 yuan/ton, with UCO costs accounting for 78% of the total cost. Since May 2024, domestic continuous SAF production plants have had profitability. However, in 2024, domestic SAF plant orders were insufficient, and there were fewer plants in continuous production, resulting in overall lower profitability. With stable increases in SAF demand in Europe, the UK, and elsewhere, domestic SAF plants may see stable production, thereby increasing the profitability of SAF production enterprises in China, benefiting domestic companies with SAF and UCO production capabilities. Beneficiary stocks: SAF: Zhejiang Jiaao Enprotech Stock (603822.SH), Beijing Haixin Energy Technology (300072.SZ), Penyao Environmental Protection (300664.SZ), and Longyan Zhuoyue New Energy (688196.SH) among others. UCO: Shandong High Speed Renewable Energy (000803.SZ), Shenzhen Lions King Hi-Tech (301305.SZ) among others. Risk warning: Policy advancement falls short of expectations, significant increases in raw material prices, technological progress falls short of expectations, etc.
21/02/2025
loading

©️2013 - 2025 GMT EIGHT Holdings. All Rights Reserved.

Contact: contact@gmteight.com