Leveraging a Language Model (LLM) architecture, the chatbot utilizes predefined rules for efficient user interaction. Key components include natural language understanding, dialogue management, and response generation. Challenges such as rule creation and handling ambiguous queries are addressed, alongside strategies for performance optimization. Experimental evaluation demonstrates the chatbot's effectiveness in providing accurate responses across various scenarios. This research significantly contributes to offline chatbot technology, presenting a viable solution for deploying intelligent systems in resource-constrained environments.