To avoid chromium and nickel series stainless steel being regarded as ordinary steel, it is necessary to perform classification when recovering stainless steel. In this study, a stainless steel fine classification model based on laser-induced breakdown spectroscopy (LIBS) and the random forest algorithm is proposed. Four grades of stainless steel 201, 304, 316, and 430 were selected as experimental samples, and 12 analytical spectral lines of Cr, Ni, Mn, Mo, and Fe were used as input characteristic quantities. The results show that the average recognition accuracy of 100 classification experiments on 300 groups of data is 98.28%, the modeling time is 0.418 s, the standard deviation is 0.20%, and the average classification time of a single group of data is only 0.019 s, indicating that the proposed classification model has good stability and efficiency. Owing to LIBS technology, the proposed model can be used in the field of online rapid classification of stainless steel.