Deep partisan conflict in the mass public threatens the stability of American democracy. We conducted a megastudy on a national sample of American partisans (n = 32,059) testing 25 interventions designed to reduce anti-democratic attitudes and partisan animosity. These interventions were selected from a pool of 252 interventions submitted by social scientists, practitioners, and activists as part of the Strengthening Democracy Challenge. Contrary to the expectations of expert forecasters, we find that nearly every selected intervention (23 out of 25) significantly reduced partisan animosity. We also identify several interventions that successfully reduced the other outcomes targeted in the Challenge – support for undemocratic practices and partisan violence – as well as a number of related secondary outcomes, including support for undemocratic candidates, opposition to bipartisan cooperation, and biased evaluations of politicized facts. Furthermore, by examining the observed pattern of effect sizes, we also gain insight into the underlying structure of these outcomes. There is little overlap between the interventions that affect partisan animosity and those that affect support for undemocratic practices or partisan violence, suggesting that these outcomes are largely driven by separate factors. However, we do find substantial overlap between the interventions that affect partisan animosity and those that affect a number of important outcomes, including biased evaluation of politicized facts, general social distrust, and preferences for social distance from outpartisans. We also found that support for undemocratic candidates was moved by interventions that affect either partisan animosity or support for undemocratic practices, suggesting two separate causal paths. Taken together, our findings provide a toolkit of promising interventions for practitioners and shed new theoretical light on challenges facing American democracy.