Using Local LLM Tools to Optimize Chinese High School Chemistry Education: Practice, Challenges, and Future Directions

This study focuses on the structural problems of teacher shortage, unequal resource allocation, and traditional teaching mode in high school chemistry education in China and explores the application potential of local large language model (LLM) tools in personalized teaching and alleviating teachers’ workload. Taking multiple schools at different educational development levels in Henan Province as samples, through stratified random sampling, a questionnaire survey, and 44 h of intervention course practice, the auxiliary effect of LLM tools in chemistry teaching was systematically evaluated. The results show that in areas with relatively weak educational resources, LLM tools can play an important role in supplementing the imbalance of the teacher–student ratio, improving the effect of after-school personalized tutoring and expanding the depth of knowledge while significantly shortening the time for teachers to prepare lessons and design lesson plans; however, in the fields of chemical equation correction, complex calculation, and professional symbol recognition, their accuracy and convenience of operation still need to be further improved. Based on the theory of digital transformation of education, this paper also discusses the necessity of technology, teachers, and policies to jointly build an intelligent education ecosystem and provides theoretical and practical references for the future construction of subject-specific LLM tools, optimization of teacher training mechanisms, and improvement of data privacy protection strategies.