A new polynomial machine learning potential (PolyMLP) model driver with various parameterizations is now available in OpenKIM (DOI: 10.25950/948ad72c). Accompanying the driver are 837 parametrizations for 167 material systems imported from the PolyMLP Repository developed by Dr. Atsuto Seko (Kyoto University). Each material system except CuAgAu has several parametrizations which all lay on a Pareto front of computational time and RMS accuracy. For information regarding the different models available, as well as to see which one is recommended as the best trade-off between speed and accuracy, see the "disclaimer" metadata field of any model for the system you're interested in.