With MYNTS-T, traffic network problems can be formulated using a linguistic approach as a translation between two domain-specific languages. The input is a graph with network properties and a set of user-defined translation rules. The output is a problem description in the form of a system of nonlinear equations suitable for generic nonlinear solvers. The translation algorithm has a computational complexity that is linear with the size of the graph and the number of equations per element. The algorithm allows element-wise separation and can be massively parallelized. The algorithm's viability and performance have been tested on several realistic examples, including gas transmission, power grids, water supply, cooling circuits, and multisector coupling.