PROCESS is a reactor systems code – it assesses the engineering and economic viability of a hypothetical fusion power station using simple models of all parts of a reactor system, from the basic plasma physics to the generation of electricity. It has been used for many years, but details of its operation have not been previously published. This paper describes some of its capabilities. PROCESS is usually used in optimisation mode, in which it finds a set of parameters that maximise (or minimise) a figure of merit chosen by the user, while being consistent with the inputs and the specified constraints. Because the user can apply all the physically relevant constraints, while allowing a large number of parameters to vary, it is in principle only necessary to run the code once to produce a self-consistent, physically plausible reactor model. The scope of PROCESS is very wide and goes well beyond reactor physics, including conversion of heat to electricity, buildings, and costs, but this paper describes only the plasma physics and magnetic field calculations.The capabilities of PROCESS in plasma physics are limited, as its main aim is to combine engineering, physics and economics. A model is described which shows the main plasma features of an inductive ITER scenario. Significant differences between the PROCESS results and the published scenario include the bootstrap current and loop voltage. The PROCESS models for these are being revised. Two new models for DEMO have been obtained. The first, DEMO A, is intended to be “conservative” in that it might be possible to build it using the technology of the near future. For example, since current drive technologies are not yet mature, only 12% of the current is assumed to be due to current drive. Consequently it is a pulsed machine, able to burn for only 1.65hours at a time. Despite the comparatively large size (major radius is 9m), the fusion power is only 1.95GW. The assumed gross thermal efficiency is 33%, giving just 465MW net electric power. The second, DEMO B, is intended to be “advanced” in that more optimistic assumptions are made. Comparison of DEMO A and B with a reference ITER scenario shows that current drive and bootstrap fraction need the most extrapolation from the perspective of plasma physics.