MANILA is a low-code web application to support the specification and execution of machine learning fairness evaluations. In particular, through MANILA it is possible to evaluate different combinations of ML models and fairness-enhancing methods combinations using several effectiveness and fairness metrics. MANILA is made of two main components: 1. A front-end application that guides the data scientist in selecting the features (i.e., ML methods, quality-enhancing methods, validation techniques, and metrics) they want to use in the evaluation process. The form provides a set of logical constraints among the features that will lead to an always executable experiment. 2. A back-end application that implements the fairness evaluation process, starting from the selected features. In particular, the evaluation process involves applying each fairness-enhancing method to each ML method and computing the selected metrics accordingly. MANILA provides two ways to execute the evaluation process: * it can generate the Python code implementing the process so that it can be executed locally or embedded in other pipelines; * it can directly execute the process in the back end, showing the results to the data scientist. Eventually, MANILA returns the ML setting (i.e., ML algorithm and fairness-enhancing method), performing better under the selected metrics.