Assessment of allergenic potency of low molecular weight compounds is generally performed using animal models, such as the guinea pig maximisation test and the murine local lymph node assay (LLNA). Progress in unravelling the mechanisms of skin sensitisation, including effects on the production of cytokines by the different cell types of the skin, provides us with the opportunity to develop in vitro tests as an alternative to in vivo sensitisation testing. The aim of the present study was to establish an in vitro method to assess the potency of allergens, on the basis of their induction of cytokine production by murine and human keratinocytes. In the present study we used test systems comprised of the murine epidermal keratinocyte cell line HEL-30 and the human keratinocyte cell line HaCaT. We exposed these cell lines to the allergens ethyl-p-aminobenzoate (benzocaine), diethylamine (DEA), 2,4-dinitrochlorobenzene (DNCB), and phthalic anhydride (PA). IL-1alpha and IL-18 dose-response data were evaluated by non-linear regression analysis and at a stimulation index of 3 of cytokine production of treatment versus control, the corresponding allergen concentration was calculated. For HEL-30, for both cytokines DNCB showed the strongest potency followed in this order by PA, benzocaine, and DEA. This classification was similar to our previous findings obtained in the LLNA. For HaCaT, unfortunately, such ranking proved to be much less feasible. In conclusion, to assess the potency of allergens the murine keratinocyte cell line HEL-30 may be a useful in vitro test system, alternative to in vivo models, although this requires further testing using a much wider range of compounds.