Imrecoxib is a novel cyclooxygenase-2 inhibitor independently developed in China, which exhibits a good efficacy and tolerance in orthopedic disorders. The current study aimed to further compare its efficacy and safety with celecoxib as postoperative analgesia in arthroscopic knee surgery (AKS). Patients receiving AKS were enrolled and randomly assigned to imrecoxib (n = 64) and celecoxib (n = 62) group to receive analgesia for 72h after surgery. Pain at rest and movement, pethidine consumption, patient's satisfaction, Lysholm score, and adverse events were assessed after AKS. Meanwhile the upper limit of 95% CI of pain-score mean difference (MD) between imrecoxib and celecoxib was calculated, then, the non-inferiority was defined if the all-time-point upper limits of 95% CI less than 1. Imrecoxib was non-inferior to celecoxib for alleviating pain at rest (upper limit of 95% CI of MD ranging from 0.443 to 0.782, all time-point values less than 1); as well as for attenuating pain at movement (upper limit of 95% CI of MD ranging from 0.398 to 0.582, all time-point values less than 1). Moreover, rescue analgesia rate (P = 0.583), pethidine consumption (P = 0.454), patient's satisfaction at 72h (P = 0.408), and Lysholm score at M3 (P = 0.776) were of no difference between imrecoxib group and celecoxib group. Additionally, the main adverse events in two groups were nausea (P = 0.425), constipation (P = 1.000), vomiting (P = 0.715), headache (P = 1.000), and dizziness (P = 0.667), which were mild and manageable. Imrecoxib is non-inferior to celecoxib in postoperative analgesia and exhibits an acceptable tolerance in patients undergoing AKS.