Context. This paper describes the initial modelling of gas and dust data acquired in August and September 2014 from the European Space Agency's Rosetta spacecraft when it was in close proximity to the nucleus of comet 67P/Churyumov-Gerasimenko. Aims. This work is an attempt to provide a self-consistent model of the innermost gas and dust coma of the comet, as constrained by the Rosetta Orbiter Spectrometer for Ion and Neutral Analysis (ROSINA) data set for the gas and by the Optical, Spectroscopic, and Infrared Remote Imaging System (OSIRIS) data set for the dust. Methods. The model uses a previously developed shape model for the nucleus, and from this the water sublimation rate and gas temperatures at the surface are computed with a simple thermal model. The gas expansion is modelled with a 3D parallel implementation of a Direct Simulation Monte Carlo algorithm. A dust drag algorithm is then used to produce dust densities in the coma, which are then converted to brightnesses using Mie theory and a line-of-sight integration. Results. We show that a purely insolation-driven model for surface outgassing does not produce a reasonable fit to ROSINA/COPS data. A stronger source in the "neck" region of the nucleus (region Hapi) is needed to match the observed modulation of the gas density in detail. This agrees with OSIRIS data, which shows that the dust emission from the "neck" was dominant in the August-September 2014 time frame. The current model matches this observation reasonably if a power index of 2-3 for the dust size distribution is used. A better match to the OSIRIS data is seen by using a single large particle size for the coma. Conclusions. We have shown possible solutions to the gas and dust distributions in the inner coma, which are consistent with ROSINA and OSIRIS data.