The stellar initial mass function (IMF) is an essential input for many astrophysical studies but only in a few cases has it been determined over the whole cluster mass range, limiting the conclusions about its nature. The 25 Orionis group (25 Ori) is an excellent laboratory for investigating the IMF across the entire mass range of the population, from planetary-mass objects to intermediate/high-mass stars. We combine new deep optical photometry with optical and near-infrared data from the literature to select 1687 member candidates covering a 1.1{deg} radius area in 25 Ori. With this sample we derived the 25 Ori system IMF from 0.012 to 13.1M_{sun}_. This system IMF is well described by a two-segment power law with {GAMMA}=-0.74+/-0.04 for m<0.4M_{sun}_ and {GAMMA}=1.50+/-0.11 for m>=0.4M_{sun}_. It is also well described over the whole mass range by a tapered power-law function with {GAMMA}=1.10+/-0.09, mp=0.31+/-0.03 and {beta}=2.11+/-0.09. The best lognormal representation of the system IMF has mc=0.31+/-0.04 and {sigma}=0.46+/-0.05 for m<1M_{sun}_. This system IMF does not present significant variations with the radii. We compared the resultant system IMF as well as the brown dwarf/star ratio of 0.16+/-0.03 that we estimated for 25 Ori with that of other stellar regions with diverse conditions and found no significant discrepancies. These results support the idea that general star-formation mechanisms are probably not strongly dependent on environmental conditions. We found that the substellar and stellar objects in 25 Ori do not have any preferential spatial distributions and confirmed that 25 Ori is a gravitationally unbound stellar association.