The Global Biodiversity Information Facility (GBIF) provides copious species occurrence records, and our submission harnesses these data to generate high-quality models that estimate a species’ abiotic niche requirements and suitable geographic areas. Although highly useful for pressing environmental issues, such models generally suffer from major obstacles, including minimizing the effects of sampling bias and evaluating performance to identify optimal model complexity. With collaborators, we recently authored two R packages (spThin and ENMeval) that automate solutions to these obstacles. For this submission, we used the R package shiny to create “Wallace: Harnessing Digital Biodiversity Data for Predictive Modeling, Fueled by R,” which integrates these packages with GBIF data via a Graphical User Interface (GUI). Researchers can download and map GBIF occurrence data, eliminate questionable records, remove clustered records, access climatic variables, build and evaluate Maxent models of varying complexities, visualize predictions, and save results. Wallace, beta version 0.1 can be run online or locally (PC/Mac), and was designed as modular and expandable. It maintains clear linkages to R packages, hopefully enticing others to add other options/packages later (e.g. databases/variables, data-cleaning, and modeling algorithms). We demonstrate Wallace’s functionality for the Andean Spectacled Bear (Tremarctos ornatus), a species of conservation concern.