Background: Bipolar disorder (BD) is a severe psychiatric disorder that is often misdiagnosed and under-diagnosed in clinical settings. The 33-item Hypomania Checklist (HCL-33) is a newly developed self-administered scale for BD detection, while the 33-item Hypomania Checklist-external assessment (HCL-33-EA) is a version of the HCL-33 for external rating used by patient's carer (e.g., family member or friend). We aimed to compare the screening abilities between the HCL-33 and the HCL-33-EA, and evaluate the screening consistency between the two scales. Methods: The data were collected from 269 patients with diagnosed BD (n = 84) or major depressive disorder (MDD) (n = 185). The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area under the curve (AUC) between the HCL-33 and the HCL-33-EA for BD were compared against clinician diagnosis as the gold standard. Results: Using Youden's index, the optimal cut-off value of the HCL-33 is 20, while the corresponding figure for HCL-33-EA is 11. Using Youden's index, the HCL-33-EA showed a better performance than the HCL-33 (0.51 vs.0.41). The HCL-33-EA was more sensitive in correctly identifying BD patients from MDD patients than the HCL-33 (0.83 vs. 0.59), while the HCL-33 presented better specificity than the HCL-33-EA (0.82 vs. 0.68). There was significant screening consistency between the two scales (p < 0.001). Conclusions: Both scales have acceptable psychometric properties in detection BD from MDD. Use of the two scales should be considered based on the assessment purpose in clinical research or daily practice (i.e., prefer sensitivity or specificity). Noticeably, the current sample size is insufficient and future studies are recommended to further evaluate the scales.