Artificial intelligence is already all around us and has been applied to almost every aspect of society and business. One of the most striking innovations in the application of AI has been the introduction of large legal language models in judicial decision-making. There has been growing interest in the use of AI in legal systems worldwide in recent years, particularly in the role of judges. The main question addressed in this Article is that what should be the potential legitimacy, weaknesses, and limitations of large legal language models in judicial scenery. To address it, it takes the Chinese smart court construction as an example and studies 133 cases of smart courts from 2017 to 2024. It summarizes four patterns from Shanghai city, Zhejiang province, Jiangsu province, and Shenzhen city. Based on this, this article analyzes the achievements and shortcomings of the application of large language models in China’s smart courts.