Abstract: With its rapid development and widespread, the Internet offers convenient and shortcut ways to retrieve information on it. However, to search such huge and quickly changing information on the Internet, more exact and faster retrieval tools should be developed. In this paper, an information retrieval server based on ontology and multi-agent is presented, which integrates several kind of agents, such as interface agent, pre-processing agent, management agent, information processing agent, and information searching agent with mobile ability. The system also uses ontologies to classify the domains of documents and assist user to normalize their queries. Using this system, dynamic changes of information on the Internet can be reflected timely, and the navigational ability of retrieving information can be improved.