Malaria is one of the most common infectious diseases worldwide, with 263 million cases and 507,000 deaths in 2023. Multiple species of Plasmodium cause malaria in humans, with P. falciparum being the most common and deadliest. Inferring genetic clusters is important for epidemiology, for identifying the specific species causing disease, and for tracking transmission.   Methods currently rely on differences in particular genetic markers, such as SNPs in antigen genes and microsatellites. There are also comparisons at the whole genome level, including studies in specific countries and the Malaria Genomic Epidemiology Network, which has compared thousands of genomes from P. falciparum and P. vivax. However, these genomics methods are commonly based on comparisons to a single reference genome. Such methods require computationally intensive bioinformatics pipelines and substantial expert knowledge, making them potentially unsuitable for resource-poor settings with the highest disease burden.   To combat this, we tested whether a reference-free method, SKA2, could be used to cluster Plasmodium genomes at different levels of diversity.  To do this, we collected a set of 89 genomes from 16 Plasmodium species infecting humans and animals, assembled at the chromosome or complete level. DNAdiff was used to establish a ground truth SNP distance between samples and then compared to SKA2 using both simulated reads and re-assembled genomes as inputs. Computational resources for each method were also assessed to determine if this pipeline is feasible in a resource limited setting.