The chatbot uses OpenAI's Chat Completion endpoint with GPT-4 to generate Cypher statements based on user input. The model is trained on a knowledge graph database, which provides complete and total control over the answers provided by the chatbot. The chatbot can answer follow-up questions that rely on information provided in the conversation history, making the whole discussion smoother. It has excellent value for answering follow-up questions and can make the user feel like they are talking to someone who understands their meaning. However, it's recommended to use GPT-4 when possible, especially for generating Cypher statements, as it is better at following instructions and understanding the given task. Additionally, the model should only return Cypher statements, but it might override system instructions and return previously provided database results in some cases. The code is available on GitHub.