On February 1, SEU marked the global debut of the “Ctogen” large model for concrete material science. This model was jointly developed by the National Key Laboratory for Major Infrastructure Engineering Materials, SEU’s Big Data Center, and Alibaba Cloud Computing Co., Ltd. It is the first large model for concrete material science in China. This achievement is centered around the deep integration of “Artificial Intelligence + Concrete Materials,” breaking the limitations of conventional engineering material development and application paradigms. It marks China’s entry into the international leading ranks in the field of intelligent engineering materials andprovides a “Chinese solution” for the digital and low-carbon transformation of global infrastructure construction.
Compared to the “hallucinations” and uncertainties of general-purpose large models in specialized fields, the “Ctogen” modelderives its core competitivenessfrom “deep specialization.” It can not only provide expert-level and traceable intelligent Q&A, offering precise answers to complex issues in concrete material research and development, design, and engineering applications but also integrate four key knowledge processing tools: intelligent data extraction, multi-dimensional data comparison, literature review generation, and knowledge graph construction. These tools compress tasks that traditionally required days or even months—such as literature review and data analysis—to a matter of minutes, significantly enhancing the productivity of researchers and engineers. Crucially, the model features an open intelligent agent platform, which includes three core agents: performance prediction, mix proportion design, and crack risk assessment. Through the collaborative synergy of “l(fā)arge model + small models,” the platform enables an end-to-end closed-loop process from parameter input to solution generation. It provides standardized, high-precision quantitative decision-making tools for complex and experience-dependent challenges such as “from performance objectives to optimal formulations” and “from structural parameters to crack risk assessment,” thereby bridging a critical technological gap in the intelligent design of concrete materials.
From the perspective of application value, the “Ctogen” large model will comprehensively empower the entire infrastructure construction chain. In terms of R&D, through its intelligent Q&A and mix proportion design agents, the model can accelerate innovation in high-performance, low-carbon concrete materials, driving the development of green building materials. During construction and operation, leveraging its crack risk assessment agent and multi-dimensional data comparison tools, it enables dynamic monitoring of engineering quality and early risk warnings, enhancing the durability of infrastructure. In the long term, the model will continuously refine the industry knowledge graph and expand its ecosystem of intelligent agents. It will provide a scalable AI-supported platform for cutting-edge fields such as service life prediction and intelligent construction, thereby facilitating the intelligent and sustainable development of infrastructure under the “carbon peaking and carbon neutrality” goals.
Liu Jiaping, Academician of the Chinese Academy of Engineering and Chief Professor of SEU stated: “This model propels concrete material research and development into the data-driven ‘fourth paradigm.’ It provides a benchmark solution for addressing critical challenges related to data, algorithms, and models in the engineering materials field, thereby accelerating the low-carbon transformation and high-quality development of the building materials industry.”
The “Ctogen” large model was officially unveiled at the 2026 First Council Meeting and 2025 Annual Academic Committee Meeting of the National Key Laboratory for Engineering Materials of Major Infrastructure Projects. As a core segment of the conference, leaders and experts including Wu Xiaocheng, Secretary of the CPC SEU Committee; Lin Tao, Standing Committee Member of the CPC Nanjing Municipal Committee and Secretary of the CPC Jiangning District Committee; Shi Xiaopeng, Deputy Director-general oftheJiangsu Provincial Department of Science and Technology; Lu Hongwei, Deputy Director-general of Jiangsu Provincial Department of Housing and Urban-Rural Development; Academicians of the Chinese Academy of Sciences Wei Bingbo and Zhang Tongyi;andAcademicians of the Chinese Academy of Engineering Nie Jianguo, Miao Changwen, Ren Huiqi, Feng Zhihai, Xing Feng, and Liu Jiaping, along with enterprise representatives, jointly unveiled the large model. They also signed a joint research agreement for intelligent agents, fostering multi-party collaboration to support the model’s future iterations and industrial implementation.As one of the supporting institutions, SEU will continue to focus on major national strategic needs, deepen interdisciplinary research on“AIand Engineering Materials,” and promote the transformation of more frontier scientific and technological achievements, contributing to the high-quality development of China's infrastructure and the realization of high-level sci-tech self-reliance.
Source: SEU News Network
Translated by: Melody Zhang
Proofread by: Gao Min
Edited by: Leah Li
