Background and Purpose: Considering the potential impact of warm ischemia time (WIT) on renal functional outcomes after robot-assisted partial nephrectomy (RAPN), many techniques that reduce or eliminate WIT have been studied. We present our institutional experience and progression using one such technique-off-clamp RAPN-as well as the results of this technique in the management of complex cases. Patients and Methods: A retrospective chart review of 65 patients undergoing off-clamp RAPN was performed, 15 of whom underwent off-clamp RAPN for 26 complex tumors. Complex features included hilar location, completely endophytic growth, and ipsilateral multifocality. In all cases, hilar vessels were dissected but not clamped. Results: Mean tumor size was 2.5 cm (standard deviation; [SD]=1.4), while mean nephrometry score was 8.7 (SD=1.5). One (7%) intraoperative complication occurred. Mean estimated blood loss was 403 mL (SD=381), mean operative time was 190 minutes (SD=68), and WIT was 0 minutes in all cases. Mean length of stay was 1.8 days (SD=0.9), with one patient needing a postoperative blood transfusion (Clavien II complication). Final pathology results demonstrated clear-cell carcinoma (n=16), papillary carcinoma (n=4), angiomyolipoma (n=1), oncocytoma (n=2), and cystic nephroma (n=3). Margins were negative for tumor for 96% (25/26) of resected masses. Estimated glomerular filtration rate (eGFR) decreased by an average of 3.1 mL/min/1.73m2 (SD=9.8, P=0.24), at a mean follow-up of 177 days (SD=296). Five patients with radiographic follow-up of at least 6 months have no evidence of disease recurrence. Conclusions: Off-clamp RAPN can be safely and effectively performed even in the case of complex tumors, but occurs with higher estimated blood loss. Minimal changes in eGFR were experienced by patients undergoing off-clamp RAPN at an average follow-up of roughly 6 months. Longer follow-up and direct comparison with conventional clamped RAPN technique are needed to establish the efficacy of off-clamp RAPN in complex cases.