This paper studies the application of large language models（ LLMs） in English reading instruction, exploring their role in enhancing students′ higher-order thinking skills and deep reading abilities. Through teaching experiments, the study conducted pretest, mid-test, and post-test on students′ reading comprehension levels. The results verify the positive impacts of LLMs on students′ language comprehension, text analysis, and critical thinking in a flipped classroom model. The findings indicate that using LLMs for English learning not only improves students′ reading comprehension levels but also strengthens their higher-order thinking abilities, thereby promoting deep learning. This research provides theoretical and practical guidance for the application of large language models in English reading instruction.