This paperdefines Pi, a parallel archiecture interface that separates model and machine issuses, allowing them to be addressesd independently. This provides greater flexibility for both themodel and machine builder. Pi addresses a set of common parallel model requirements including lawa latency communications, fast task switching, low cost synchronization, efficient, storage management, the ability to exploit locality, and efficeient support for sequential code. Since Pi provides generic parallel operations, it can efficiently support many parallel programming models including hybrids of existing models. Pi also forms a basis of comparison for architectural componenets. This pater presents an overview of Pi, plus a decription of several model examples wiich have been constructed and evaluated on the interface.