The Canahuire epithermal Au-Cu-Ag deposit in southern Peru is a complex and challenging deposit for metallurgical studies. It has issues of gold locking in sulfides and non-sulfides, deleterious elements and acid rock drainage (ARD). From scoping study to feasibility study, there has been considerable improvement in the understanding of the important geometallurgical characteristics that need to be measured and modelled. This paper presents the current understanding and the requirements to continue to provide optimal models for mining studies.Significant effort has been put into understanding the genesis of the deposit. Improved spatial and temporal understanding of lithology, alteration, structure and mineralogy was used to build a geological and alteration model of the deposit.Early metallurgical programs quantified the effect of head grade on process plant recoveries, but the geometallurgical and statistical analyses did not identify other physical characteristics that influence these recoveries. The studies improved the understanding of the nature of locked gold which can occur as minute inclusions (< 5 m) in sulfides (primarily pyrite with lesser arsenopyrite, pyrite being 100 times more abundant than arsenopyrite) and gangue minerals (quartz and carbonates) as well as solid solution in arsenopyrite and pyrite.The presence of pyrite and the limited buffering capacity of waste rocks increases the potential ARD, which will require mitigation costs before, during and after operations. Waste rock ARD potential has been classified and incorporated in the geometallurgical model. The current classification of waste rock at Canahuire, using international classification standards, is limited due to the complexity of the deposit.Complex deposit metallurgy necessitates a complex process flow sheet, including flotation and cyanidation. Economic recovery will be constrained by the relationships between economic variable grade, other elements and mineralogy. These constraints and inter-relationships are difficult to establish throughout the deposit with a single test. A follow-up geometallurgical program is under investigation and will attempt, with the help of additional test work, to map areas of higher and lower recoveries. It is hoped that metallurgical response can be linked to measurable parameters other than head grade. If this is possible, these factors can be incorporated into future geometallurgical models.CITATION:Baumgartner, R, Dusci, M, Trueman, A, Brittan, M and Poos, S, 2013. Building a geometallurgical model for the Canahuire Epithermal Au-Cu-Ag Deposit, Southern Peru - past, present and future, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 51-58 (The Australasian Institute of Mining and Metallurgy: Melbourne).