Virtual screening techniques have gained much attention as a means of studying bioactive peptides. This study aimed to screen DPP-IV inhibitor peptides in goat milk after simulated digestion in vitro combined with molecular docking and dynamics simulations. By evaluating the docking energy and active sites, and by analyzing RMSD, RMSF, and Rg values, two novel peptides, GPFPLL and LPYPY, were successfully screened and identified. GPFPLL and LPYPY were found to exhibit high inhibitory activity against DPP-IV (IC(50) of 130.68 ± 10.38 μM and 179.52 ± 18.89 μM, respectively). Both GPFPLL and LPYPY stably bound to S1 and S1' in DPP-IV, and both demonstrated competitive inhibition of DPP-IV. The inhibition of DPP-IV by GPFPLL and LPYPY after in vitro digestion reached 31.90 % ± 1.80 % and 39.37 % ± 0.90 %, respectively. In a Caco-2 cell experiment, GPFPLL and LPYPY exhibited significant inhibition of DPP-IV, reaching 46.53 % ± 3.48 % and 65.98 % ± 2.87 %, respectively, when the concentration of each peptide was 2 mg/mL. The results of this study suggest that using molecular docking and dynamics simulations to screen novel peptides is an effective approach, and the identified peptides GPFPLL and LPYPY show potential for diabetes management.