Aspergillus terreus is widely used for industrial production of important chemical compounds such as itaconate and lovastatin. To improve understanding of its physiology, we reconstructed a genome-scale metabolic model, iJL1454, consisting of 1454 genes, 1451 reactions and 1155 metabolites, based on genome annotation and literature mining. The model accurately predicted the growth phenotype of A. terreus on different carbon and nitrogen sources, and after deletion of essential genes. In iJL1454, 69 and 20 (a subset of the 69) genes were identified as growth essential in minimal and rich mediums, respectively. Ten of the genes had previously been suggested as essential by experimental observations. The biosynthetic pathway of itaconate from glucose, in particular the coding genes and the transporters, was elucidated. The pathway included 29 reactions and 55 genes. To guide metabolic engineering, we investigated bottleneck reactions in itaconate production and the effect of perturbations of nutrient and environmental conditions. The results illustrate the value of model iJL1454 as a tool to understand and optimize A. terreus metabolism.