FlexiGait was compared against a Vicon marker-based motion capture reference in 31 participants from the public GPJATK dataset. Mean absolute error for joint kinematics was below 5°. Agreement was excellent for stride time (ICC = 0.97) and cadence (ICC = 0.96), and moderate for swing phase (ICC = 0.72). Bland–Altman analysis showed minimal bias, with a mean difference of approximately 0.01 s for step time. These results were obtained in healthy adults recorded in a laboratory setting; performance has not been established in any patient population. The study was carried out with Prof. Ali Kiapour and Prof. Santiago A. Lozano Calderon. Results were presented at the GAMMA Congress 2025 and the congress abstract is published as Safarzadeh O, FlexiGait: 3D Markerless Gait Analysis using single camera RGB videos, Gait & Posture 2025;117(Suppl 1):S29, doi:10.1016/j.gaitpost.2025.01.073. The full manuscript is available as a preprint on SSRN and has not yet completed peer review. The named investigators contributed in their individual capacity. The views represented are not official ones of Harvard University, Harvard Medical School, Massachusetts General Hospital, or any of their units or officers, and nothing here constitutes endorsement of FlexiTrace or its products by any institution.