In this paper we test how efficiently state-of-the art solvers are capable of solving credulous and sceptical argument-acceptance for lower-order extensions. As our benchmark we consider two different random graph-models to obtain random Abstract Argumentation Frameworks with small-world characteristics: Kleinberg and Watt-Strogatz. We test two reasoners, i.e., ConArg2 and dynPARTIX, on such benchmark, by comparing their performance on NP/co-NP-complete decision problems related to argument acceptance in admissible, complete, and stable semantics.