Based on the heat-clearing and detoxifying effects of Gentianae Radix et Rhizoma, the network pharmacology is mainly used to predict the potential targets of Gentianae Radix et Rhizoma for anti-inflammatory activity and to perform the experimental verification. A method for detecting the biological potency of Gentianae Radix et Rhizoma based on verifiable targets has been established to provide a reference for improving the quality evaluation and control standards of Gentianae Radix et Rhizoma. High performance liquid chromatography can be used to construct chemical fingerprints of different batches of Gentianae Radix et Rhizoma. Constructing a component-target-disease network of Gentianae Radix et Rhizoma for its anti-inflammatory activity was applied to screen potential anti-inflammatory components and related targets of Gentianae Radix et Rhizoma, and to verify the target of Gentianae Radix et Rhizoma by using biological evaluation methods. Detecting the biological potency of different batches of Gentianae Radix et Rhizoma extracts was used to inhibit COX-2 enzyme activity based the verifiable target cyclooxygenase-2(COX-2). The results showed that different batches of Gentianae Radix et Rhizoma accorded with the pharmacopoeia testing regulations, and the chemical fingerprints have a high similarity(similarity>0.93), suggesting that there is no significant difference in the characteristics of the chemical components. Based on network pharmacology predictions, 18 candidate targets were found to have potential direct interactions with the ingredients in Gentianae Radix et Rhizoma. Among them, the most important target is COX-2. Based on the experimental verification of recombinant human COX-2 protease activity inhibition, Gentianae Radix et Rhizoma can inhibit the COX-2 enzyme activity in a dose-dependent manner. It can function with a low concentration(0.75 mg·mL~(-1)), which preliminarily confirmed the accuracy of network pharmacology prediction. The biological potency detection method of Gentianae Radix et Rhizoma based on COX-2 inhibitory activity was optimized and established. The qualitative response parallel line method was used to calculate the biological potency of anti-inflammatory activity, which ranged from 23.04 to 46.60 U·mg~(-1). For network pharmacology prediction, it can screen and clarify the possible targets of traditional Chinese medicine rapidly, which can guide the establishment of a biological evaluation method for the quality of medicinal materials with related activities. Compared with chemical fingerprints, the biological potency testing can better detect quality fluctuations of traditional Chinese medicine.