We present a receptor-modeling concept based on multidimensional QSAR (mQSAR) developed at our laboratory for the in silico prediction of the toxic potential of drugs and environmental chemicals. Presently, the VirtualToxLab includes nine so-called virtual test kits for the estrogen (alpha/beta), androgen, thyroid (alpha/beta), glucocorticoid, aryl hydrocarbon, and peroxisome proliferator-activated receptor gamma, as well as for the enzyme cytochrome P450 3A4. The surrogates have been tested against a total of 798 compounds and are able to predict the binding affinity close to the experimental uncertainty, with only six of the 188 test compounds being calculated more than a factor of 10 off the experimental binding affinity and the maximal individual deviation not exceeding a factor of 15. These results suggest that our approach is suited for the in silico identification of adverse effects triggered by drugs and environmental chemicals. In this account, we summarise the current evaluation status of the models and introduce an Internet access portal, immediately available to selected laboratories, and aimed at a peer evaluation of our concept.