Which Large Language Model should You Use in Vietnamese Education: ChatGPT, Bing Chat, or Bard?

This paper aims to determine the most suitable large language model (LLM) among ChatGPT, BingChat, and Bard for Vietnamese educators and students. The results show that ChatGPT and BingChat exhibit similar performances and outperform Bard in mathematics. In literature, ChatGPT performs better than BingChat and Bard. However, BingChat demonstrates superior performance compared to ChatGPT and Bard in English. All three LLMs demonstrate relatively comparable performances in physics. For biology, BingChat and Bard outperform ChatGPT. Notably, Bard outperforms both ChatGPT and BingChat in chemistry. In the domain of social sciences, both BingChat and Bard exhibit comparable and better performances than ChatGPT. These findings provide valuable insights for educators and students in Vietnamese high schools in selecting suitable LLM for specific subjects.