Neuropathic pain may be due to a primary insult to the peripheral or central nervous system. In this respect hyperalgesia and allodynia are the results of prostaglandin release in the spinal cord. It seems that at earlier stages of injury, COX1 release is more than COX2 isoenzyme. Comparing selective COX2 inhibitors, nimesulide, a highly selective COX2 inhibitor effectively reduces hyperalgesia due to peripheral administration of inflammatory agents like formalin. In this study, the effect of nimesulide and celecoxib was investigated in a neuropathic model of hyperalgesia. For this purpose, male Sprague-Dawley rats (n=18, 150-200 g) were divided into 3 groups, i.e., left ligation (LL, control), Sham-operated, LL+ mentioned drugs. For induction of neuropathic pain, Bennet & Xie (1988) model was used. Nimesulide (1 .25, 2.5. and 5 mg/Kg) and celecoxib (12.7 mg/kg) were used in treated groups. For thermal hyperalgesia 42 °C and 100° C water was used. In addition, acetone and Von Frey Filaments tests were used for cold and mechanical allodynia respectively. Experiments were performed one day before and 1, 3, 5, 7, 10, and 14 days post-injury. For statistical analysis, ANOVA and Tukey"s tests were used and p<0.05 was regarded as significant.The results showed that nimesulide at doses of 2.5 and 5 mg/kg reduce hyperalgesia and allodynia as compared to LL control group. There were also no significant differences between sham and LL+drugs groups. The effect of celecoxib at a dose of 12.7 mg/kg in reducing hyperalgesia as compared to LL control and nimesulide-treated group at a dose of 2.5 mg/kg began later and celecoxib had no anti-allodynic effects in this respect.It seems that nimesulide was more effective than celecoxib in reducing hyperalgesia and allodynia in neuropathic pain condition. In addition, compared to results for chronic inflammatory pain due to formalin administration, nimesuIide was more effective in controlling pain-ensued behavior.