Abstract: With the frequent occurrence of emergencies such as natural disasters, accidents, public health events, and social security incidents,etc. LLM enabled intelligent emergency response has become a frontier and hot topic in the field of emergency management disciplines. With its powerful capabilities in information collection, natural language understanding, generation, and reasoning,etc. large language models can provide scientific evidence for emergency management departments in the prevention, preparation, response, and recovery of emergencies, and can playing an important role in the field of emergency. Based on the characteristics of large language models, an emergency management framework enabled by large language model technology from two dimensions: event types and emergency management activities. Guided by the framework, the ways and methods are explored for large language model technology to enable emergency management activities for different types of events, including the identification of potential risks in the prevention stage, the generation of emergency response plans and the monitoring and identification of events in the preparation stage, the decision-making of emergency command in the response stage, and the review and summary of the emergency process in the recovery stage. The emergency management framework enabled by large language model technology provides a construction idea for intelligent emergency response.