针对教育教学中知识传递效率低与个性化指导缺失问题，提出基于智谱清言大模型的个性化学习支持系统模型，从导学问题智能生成、多模态答疑交互、学情可视化分析等环节介绍具体的创新解决方案和教学案例，通过具体实践说明该模型能显著提升师生双主体在预习、授课、评价全流程的协同效率，为教育大模型落地提供既保障技术可靠性又符合教学规律的参考路径。

Promoting artificial intelligence assisted education reform is an important component of the plan to build an education powerhouse. In response to the low efficiency of knowledge transmission and lack of personalized guidance in education and teaching, this paper studies personalized learning support technology based on large-scale language models. The main content includes the following two points:(1) By constructing a dynamic reasoning mechanism guided by knowledge graphs, the engineering of prompt words is deeply coupled with the cognitive logic of the subject, effectively alleviating the problem of "illusion" in large models; Adopting a hierarchical teaching architecture to achieve collaboration between teacher experience and AI generated content, forming an interpretable intelligent teaching assistance system.(2) A personalized learning support system model based on ChatGLM-6B model is proposed. Innovative solutions and teaching cases have been proposed in the intelligent generation of guided questions, multi-modal Q&A interactions, and visual analysis of learning contexts. This can significantly improve the collaboration efficiency between teachers and students throughout the entire process of preschool education, teaching, and evaluation, providing a reference path for implementing an educational model that ensures both technological reliability and conforms to teaching laws.