This study aims to analyze the differences in improving students' systems thinking skills between guided inquiry and structured inquiry learning models. This research is a quasi-experimental research with Nonequivalent Control Group Design. The research subjects were 72 students from a high school in the city of Bandung. The research instrument used was a limited description test made based on indicators of system thinking skills that were measured including examining the components in the system; check the function of each component in the system; analyze the relationship of each component in the system; analyze the relationship between one system and another system; and analyze the balance in the system. The N-Gain test was used to analyze the increase in students' systems thinking skills in both inquiry classes and inferential statistical tests in the form of the Mann-Whitney nonparametric test to determine differences in students' systems thinking abilities in the two inquiry classes. Improving students' systems thinking abilities in the guided inquiry class obtained an N-Gain value of 0.73 with high criteria, while in the structured inquiry class an N-Gain value of 0.56 was obtained with medium criteria. The N Gain value is an indicator analyzing the relationship of each component in the system. Based on the results of inferential statistical analysis, it shows that there are significant differences in students' systems thinking abilities between guided inquiry classes and structured inquiry classes, in which students' systems thinking abilities in guided inquiry classes are better than structured inquiry classes.