The attractive properties of carbon-based nanoparticles such as graphene and its derivatives or carbon nanotubes lead to their use in many application fields, whether they are raw or functionalized, such as oxidized. These particles may finally contaminate the aquatic compartment, which is a major receptacle of pollutants. The study of their impact on aquatic organisms is thus essential. At the nano scale, recent studies have highlighted that specific surface area should be used as the most relevant descriptor of toxicity instead of the conventional mass concentration. By using a dose-response model, this work compares the chronic toxicity observed on Xenopus laevis larvae after 12-day in vivo exposure to raw, oxidized carbon allotropes, or in the presence of chemical dispersant. We show that chemical dispersion does not influence the observed chronic toxicity, whether it is through surface chemistry (oxidation state) or through the addition of a dispersant. The biological hypothesis leading to growth inhibition are discussed. Finally, these results confirm that surface area is the more suited metric unit describing growth inhibition.