Background. Topical nonsteroidal anti-inflammatory drugs (NSAIDs) are widely prescribed for the treatment of symptomatic osteoarthritis (OA) of the knee. However, searching for alternatives such as locally available medicinal herbs to manage OA knee pain remains of clinical value. The objective of the present study was to evaluate the efficacy and safety of two yellow oil formulations in patients with OA of the knee. Methods. This prospective, randomized, single-blind, active-controlled, noninferiority study enrolled 102 patients with OA of the knee. Eligible patients were randomly assigned to apply either yellow oil formulation 3 (YOF3), yellow oil formulation 4 (YOF4), or indomethacin solution (INDO) topically four times daily for four weeks. Outcomes were assessed on a biweekly basis. The primary efficacy outcome measure was a 100 mm visual analog scale (VAS) of pain, while secondary endpoints included knee function, physical performance assessments, and safety parameters. Modified intention-to-treat and per-protocol analyses were applied. Assessment of noninferiority was done with a prespecified margin of 10 mm for VAS pain. Results. Of 102 patients enrolled, 86 completed the study: 29/34 in the YOF3 group, 25/34 in the YOF4 group, and 32/34 in the INDO group. The absolute reduction in VAS pain at the final evaluation was −25.06 ± 13.91, −18.50 ± 16.06, and −23.38 ± 10.05 mm in the YOF3, YOF4, and INDO groups, respectively (p=0.169). Only YOF3 was found to be noninferior to INDO. Other efficacy outcomes were significantly improved in all three groups. All the interventions were well tolerated; no skin rash was observed in any of the three groups. Conclusions. YOF3 was shown to be noninferior to INDO in relieving knee pain and should be considered an alternative for the treatment of symptomatic OA of the knee. Further research into the mechanism of action of YOF3 and its long-term efficacy and safety is required.