As COVID-19 continues to spread around the globe, it is critical to understand the true burden of a future outbreak in developing countries like Colombia where data may be limited. Here, we estimated the rate of the initial exponential growth of cases and the basic reproductive rate for the disease. We use models with different modeling assumptions to study the differences between five major Colombian cities and between selected Latin American countries. Using an ensemble modeling technique, we estimated that the reproduction number in Colombia varied from 1.10 in Cartagena to 1.75 in Medellin with Cali being 1.47. In Latin America, Ecuador has highest initial epidemic growth rate and Panama the lowest with Colombia in middle of the list. The choice of appropriate model and parameter estimates for a location provided different scenarios in outbreaks. This analysis provides a framework for the decision makers to be better prepared for an outbreak.