As a typical difficult-to-machine material， cemented carbide YG10 was prone to cause severe tool wear when common cutting method was used.In response to this problem，laser-assisted cutting method was proposed for machining.By comparing the tool wear conditions under the two machining methods of ordinary cutting and laser-assisted cutting，it was demonstrated that laser-assisted cutting could effectively reduce cutting force and tool wear.The support vector regression model （SVR） and cross-validation-support vector regression model（CV-SVR）were established，and the amount of flank wear under specific cutting conditions were predicted.The result shows that the prediction results of the two models have a small error with the actual values，in particular，the CV-SVR model has higher fitting accuracy，compared with the SVR model，the average relative error is reduces by about 10%.The CV-optimized SVR model can effectively simulate the nonlinear relationship in tool wear，and it can provide a basis for the judgment of tool wear in actual machining.