Abstract Purpose Virtual surgical planning (VSP) has improved outcomes in bimaxillary orthognathic surgery when compared to traditional cephalometric analysis and model surgery. However, several studies have shown discrepancies between the planned and post-operative position of the maxilla with the use of VSP and splints. Patient-specific Implants (PSIs) have the maxillary movement and final position built into them as dictated by the virtual surgical plan, completely independent of the mandibular position and have shown promising results, but are limited in number and by sample size and no study to date has directly compared the accuracy of splints versus PSIs. The purpose of our study was to determine the accuracy of splints versus PSIs in executing a virtual surgical plan. Materials and Methods A retrospective analysis of patients who underwent bimaxillary orthognathic surgery with splints or PSIs planned with VSP at our institution from 2017 to present was completed. The difference in maxillary positions from the VSP to the post-operative CBCT were evaluated in both cohorts. The primary predictor variable was the planned 3D position of the maxillary mid-incisors, and right and left first molars in the antero-posterior, transverse, and vertical dimensions. The primary outcome variable was the post-operative 3D position of these points. The root mean square deviation (RMSD) was used to calculate the absolute value of the mean differences. A paired t test was used to determine if there were statistically significant differences between the predicted and observed positions with p < 0.05 Results A total of 65 patients (46 female, 19 male) aged 14 to 58 years (mean age, 23.3 years). 10 patients (8 female, 2 male) aged 14 to 54 years (mean age, 24.8 years) underwent bimaxillary surgery with a Le Fort 1 osteotomy and a mandibular sagittal split osteotomy or intraoral vertical ramus osteotomy and were rigidly fixated after the use of a splint to facilitate the final maxillary position, whereas 55 patients (38 female, 17 male) aged 14 to 58 years (mean age, 22.7 years) were rigidly fixated with the use of a PSI. The difference between the planned and actual movements varied in all dimensions in the splint cohort but was greatest in the anteroposterior dimension for all points in the PSI cohort. The only statistically significant differences in pre-operative and post-operative position of the splint group were the mid-incisor transverse and vertical dimensions, whereas all differences in the PSI cohort were found to be statistically significant with the exception of the maxillary right 1st molar in the transverse dimension and the maxillary left 1st molar in the vertical dimension. There was an overall tendency for undertreatment in all directions as shown in the deviation from 0 toward under (-) or over (+) correction, regardless of type of surgery performed. Also, there was no indication that posterior or anterior movements were more or less accurate in either group. The PSI group was more accurate at all 3 maxillary points in all 3 dimensions, ranging from 1% to 62.8% more accurate than the splint group. Conclusion The use of a PSI allows for greater control of the post-operative position of the maxilla when planned via VSP and is more accurate than the use of a splint.