Recently an approximately three year project to revise the New Zealand National Seismic Hazard Model (NSHM) was completed. This was the most significant revision of the NSHM in more than 20 years and included fundamental changes to all components of the model. An underlying philosophy of the model development was that to best represent what we know about earthquake occurrence and, hence, make the best forecast, we need to represent a wide range of datasets, hypotheses and models in all components of the NSHM. The Seismicity Rate Model (SRM) is the collection of component models that each forecast the magnitude, location and rate of earthquakes for the next 100 years. The Ground Motion Characterisation Model (GMCM) is the collection of models that forecast the range of shaking for each of the ruptures in the SRM and is covered in detail in another presentation. Broadly the SRM is broken into two components: 1) ruptures on known faults, and 2) ruptures on faults that are not yet known about. For the known faults we have implemented the UCERF inversion recipe which allows for jointly fitting multiple datasets and models to provide rates on ruptures. Some key inputs are: the Community Fault Model; Deformation models, which provide slip rates on all faults; Rupture sets, which provide geometric constraints on potential ruptures; Models of timings of past earthquakes on known faults; and, Magnitude-frequency distributions of earthquake occurrence. For unknown faults, we have developed a hybrid model that represents a significant departure from smoothed seismicity models which are typically used in seismic hazard. Similar to the inversion model, the hybrid model also combines, geological and geodetic data with the earthquake catalogue to provide a more complete forecast than smoothed seismicity alone. Models have also been developed for lower-seismicity regions which incorporate geodetic strain and the low-bias seen in the NZ earthquake catalogue in lower rate regions. Finally, the SRM accounts for the much greater variability in rate observed in New Zealand than is modelled by standard Poisson assumptions.