This paper presents the development and training of a transformer-based model using a Chinese news dataset and Chinese short stories for the task of sentence completion. The transformer model has shown remarkable success in natural language processing tasks, and this study aims to leverage its power to improve sentence completion accuracy in the Chinese language. The Chinese news dataset utilized in this research encompasses a wide range of topics, providing the model with comprehensive domain knowledge. Experimental results demonstrate the effectiveness of the transformer model on the sentence completion task in Chinese. All Codes, models, and the dataset are available at https://github.com/zhaaaos/chinese_sentence_completion