The purpose of this research is to: 1) identify differences in poetry writing skills between students taught in the conceptual writing learning model and the round table learning model; (2) knowledge of differences in poetry writing skills between high-lexical and general-lexical students. (3) knowledge of the interaction between learning models and acquisition of vocabulary for poetry skills; The study was quantitatively conducted using an empirical approach, using 2x2 factorial design, a test-form data collection technique. Prerequisite tests for data analysis consist of normality, homogeneity, and equilibrium tests. Data analysis techniques were performed using two-way analysis of variance. The conclusions drawn from this study are: (1) There is a difference in students' ability to write poetry taught through concept sentences and round table learning models (F<sub>A</sub> = 9,52 &gt; 4.01). (2) There is a difference in poetry writing skills between high-vocabulary and low-vocabulary students (F<sub>B</sub> = 35,52 &gt; 4.01). (3) There is no interaction between learning models and vocabulary acquisition with respect to students' writing skills (F<sub>AB</sub> = 0,9 &lt; 4.01).