In order to explore the influence of steel slags of different fineness on the rheological properties of the steel slag-mineral powder base polymer, the variation law of the rheology of different steel slag fineness was studied by fluidity, rheometer and water film thickness tests, and the rheological parameter prediction model was established by BP neural network based on the correlation between fluidity and WFT and rheological parameters. The results show that with the increase of steel slag fineness, the fluidity of the freshly mixed polymer mortar is improved, and its setting time and WFT are shortened. The change in slag fineness does not change the fluid type and the rheological characteristics are in line with the Bingham model. The apparent viscosity gradually decreases with the increase of shear rate, the yield stress, plastic viscosity and thixotropy continue to decrease with the increase of steel slag fineness, the rheological parameters increased with the increase of the standing time, and the growth rate also showed a gradual increasing trend. Meantime, the fluidity is positively correlated with the yield stress, and the WFT has a good linear relationship with the yield stress and fluidity. The BP neural network rheological parameter prediction model was established, and the prediction results were in good agreement and high accuracy.