Foodomics has been defined as a global discipline in which advanced analytical techniques and bioinformatics are combined to address different questions in food science and nutrition. There is a growing number of works on the development and application of non-targeted omics methods in foodomics, which reflects that this emerging discipline is already considered by the scientific community to be a valuable approach to assess food safety, quality, and traceability as well as for the study of the links between food and health. As a result, there is a clear need for more rapid, high-throughput MS approaches for developing and applying non-targeted studies. Nowadays, direct MS analysis is one of the main choices to achieve high throughput, generating a set of information from the largest possible number of samples in a fast and straightforward way. The use of high- and ultrahigh-resolution MS greatly improves the analytical performance and offers a good combination of selectivity and sensitivity. By using a range of methods for direct sample introduction/desorption/ionization, high-throughput and non-target analysis of a variety of samples can be obtained in a few seconds by HRMS analysis. In this review, a general overview is presented of the main characteristics of direct HRMS-based approaches and their principal applications in foodomics.