The tool orientation of a flat-end cutter, determined by the lead and tilt angles of the cutter, can be optimized to increase the machining strip width. However, few studies focus on the effects of tool orientation on the five-axis milling process stability with flat-end cutters. Stability prediction starts with cutting force prediction, and the cutting force prediction is affected by the cutter-workpiece engagement (CWE). The engagement geometries occur between the flat-end cutter and the in-process workpiece are complicated in five-axis milling, making the stability analysis for five-axis flat-end milling difficult. The robust discrete vector method is adopted to identify the CWE for flat-end millings, and it can be extended to apply to general cutter millings. The milling system is then modelled as a two-degrees-of-freedom spring-mass-da mper system with the predicted cutting forces. Thereafter, a general formulation for the dynamic milling system is developed considering the regenerative effect and the mode coupling effect simultaneously. Finally, an enhanced numerical integration method is developed to predict the stability limits in flat-end milling with different tool orientations. Effectiveness of the strategy is validated by conducting experiments on five-axis flat-end milling.