Gene flow is a common phenomenon for crop species and its implications for Genetically Modified Plants have raised new concerns. Undesirable effects related to gene flow may result in ecological or agronomic considerations (persistence of feral plants, creation of new weeds; impacts on biodiversity) as well as in commercial considerations (adventitious presence of GMOs in conventional crop production affecting its competitiveness in the marketplace). Consequently, the coexistence between different types of crops has become a major issue and has to be addressed per se whatever are the actual ecological, agronomic and safety impacts. On-farm gene flow occurs both in time and in space, through pollen flow as well as through seed dispersal. Several factors are involved: crop biology, landscape fragmentation, environmental conditions, crop management and post-harvesting practices. For helping in the elaboration of co-existence rules, for assessing their feasibility and their consequences as well as for setting up monitoring and control schemes, one should be able to forecast the fate of GM plants at the landscape level in the wide range of agro-ecosystems. Specific field experiments are necessary for understanding the basic phenomena involved but are difficult to extrapolate for such a perspective even if several studies have been carried out in order to broaden the scope of the evaluation: the inter-institute platforms in France or the Farm Scale Evaluation programme in the UK. For addressing such a challenge, modelling is a key element. Models reproduce the functioning of agro-systems and take into account the relevant factors and processes as well as their interactions. They thus allow simulating the behaviour of various agro-systems in non-observed situations and on a long term basis.