Core Insights - The article highlights the launch of TPT2, a significant upgrade of the time-series pre-trained transformer model by Zhongkong Technology, marking its strategic shift from an automation product supplier to an industrial AI platform provider [1][4][9] - TPT2 has received strong market interest, with 518 pre-orders and 112 transactions, indicating a robust demand for AI in industrial applications [1][4] - The model is designed to address core needs in process industries, showcasing the untapped potential of industrial AI [1][6] Group 1: Product and Technology - TPT2 is the world's first AI model deeply integrated with the first principles of process industries, utilizing a mixed expert model architecture [1][11] - The platform aims to transform the traditional project-based delivery model into a subscription service, enhancing long-term revenue stability [3][4] - TPT2's application has already been implemented in over 110 projects across various sectors, including petrochemicals and power [3][4] Group 2: Financial Performance - Zhongkong Technology reported a revenue of 9.139 billion yuan for 2024, a 6.02% increase year-on-year, with a net profit of 1.117 billion yuan, reflecting a 1.38% growth [6][7] - The industrial software revenue surged by 20.70% to 2.653 billion yuan, significantly outpacing traditional control business growth [7][8] - The gross margin for industrial software reached 54.03% in the first half of 2025, up from 31.67% in the same period of 2024, indicating effective conversion of AI technology into profitability [7][8] Group 3: Market and Strategic Positioning - The Chinese government's "Artificial Intelligence+" initiative emphasizes the integration of AI and industrial internet as a key driver for new industrialization [9][12] - Zhongkong Technology's TPT2 aligns with national strategies to cultivate intelligent native enterprises and achieve comprehensive intelligence [9][12] - The company is expanding globally, with successful bids in various international projects, showcasing its competitive edge in the global market [12][13] Group 4: Future Outlook - The industrial AI market is poised for significant growth, with TPT2 potentially unlocking a market space of 33 billion yuan annually [8][9] - Zhongkong Technology aims to transition from a product supplier to a platform-based company, enhancing its value proposition in the industrial AI sector [7][8] - The shift towards a subscription-based model is expected to drive sustainable business growth and reshape the economic landscape of industrial AI [3][4]

Core Insights - Company reported a revenue of 3.83 billion yuan and a net profit of 354 million yuan for the first half of 2025, indicating a focus on emerging businesses like industrial AI and robotics [1] - The company is advancing its "All in AI" strategy, integrating AI capabilities into its product and technology systems, and has made significant progress in the development and application of its TPT model [2] - The launch of TPT2, a platform for generating industrial agents, marks a significant upgrade, enhancing the efficiency of developing and implementing intelligent applications [3] - The robotics segment is also showing positive growth, with successful deployments of new inspection robots and significant project wins in the industrial robotics market [3] - The company is actively expanding its international presence, securing key projects in various global markets, which supports its long-term growth strategy [4] - The company has built a competitive edge through extensive data assets and industry knowledge accumulated over 30 years, which is crucial for the development of industrial AI models [5] - The company maintains the top market share in the domestic distributed control system (DCS) sector for 14 consecutive years, with a broad customer base and significant data resources supporting its AI initiatives [6] Group 1 - Company achieved a revenue of 3.83 billion yuan and a net profit of 354 million yuan for H1 2025, focusing on industrial AI and robotics [1] - The "All in AI" strategy is being deepened, with advancements in the TPT model for industrial applications [2] - TPT2 platform enhances the development and implementation of industrial solutions through natural language commands [3] Group 2 - Robotics business is expanding with successful deployments and project wins, enhancing market competitiveness [3] - International expansion is underway, with significant projects secured in various regions, supporting long-term growth [4] - The company has a competitive advantage due to extensive data assets and industry experience, essential for AI model development [5] Group 3 - The company holds the top market share in DCS for 14 years, with a wide customer base and substantial data resources for AI applications [6]

Core Insights - Zhongkong Technology has launched the TPT 2 time series model, which is based on the MoE (Mixture of Experts) framework, aimed at enhancing industrial intelligence [1] - The TPT 2 model enables closed-loop applications in various production processes through time series data algorithms, helping users ensure production safety and reduce energy and material consumption [1] - The platform shifts from a traditional "one scenario, one model" development approach to an efficient generation of intelligent agents applicable to various industrial setups through language interaction [1] Technology and Innovation - TPT 2 integrates multiple technological systems, advancing industrial applications from traditional tool-based assistants to intelligent agents capable of proactive anomaly detection, risk assessment, and autonomous decision-making [1] - The model replaces various industrial software functions such as simulation, control, optimization, prediction, evaluation, and statistics [1] Industry Collaboration - Zhongkong Technology has established the "Industrial AI Data Alliance" in collaboration with over 130 leading industry enterprises, design institutes, general contractors, and service providers [1] - The alliance aims to promote the scenario-based and large-scale application of industrial data, addressing industry challenges related to data immobility, underutilization, and insufficient collaboration [1]