In this paper, the support vector machine (SVM) is applied for the early warning of financial crisis of listed companies, financial early-warning indexes and quantitative methods are analyzed, the early-warning ability of SVM model is verified by combining with the financial data of listed companies, and the real evidence has proved the feasibility of SVM model used for financial early-warning, finally, the possibility of using multi-mode classification to identify the alert degree of financial early-warning.