We all know more and more fake video footage is popping up online every day. Models that can generate DeepFakes are rapidly maturing, creating increasingly realistic output at alarming scale. The FakeFinder project created a framework to easily experiment with the models that are successfully identifying fake videos – specifically, the top five performers from the Facebook’s DeepFake Detection Challenge in 2019. The Fakefinder platform leverages these models to batch process videos and output predictions of a video’s provenance. The tool displays individual model output so that users can compare or contrast individual model predictions and performance.