The increasing use of Nanoparticles (NPs) in a wide range of applications has led to a rising concern on the possible toxicological effects that this use may have on human health and the environment. Since experimental toxicity evaluation for the different types of NPs already available, is often expensive and time consuming, several computational approaches are proposed for the risk assessment of NPs. In this work, we have developed a predictive classification model for the toxicological assessment of iron oxide NPs with different core, coating and surface modification based on a number of different properties including size, relaxivities, zeta potential and type of coating. The model was fully validated based on several validation measurements and was released online via Enalos InSilicoNano Platform (http://enalos.insilicotox.com/QNAR_IronOxide_Toxicity/). The developed web service gives the interested user the opportunity to insert the indicated properties and get a toxicity prediction accompanied by an indication of its reliability based on the domain of applicability. This newly introduced web service complements our previously reported efforts to extract important information from available datasets and develop user friendly applications for the toxicity assessment of NPs.