Multiplicative gain modulations change the slope of the input-output relationship of neurons and alter their sensitivity to changes in excitatory inputs. They are observed frequently in vivo and are thought to play an important role for the processing of information in the brain [1]. A common feature of conditions where multiplicative modulation takes place is the presence of noise, either as background or in the driving inputs themselves [2-4]. As the amount of noise depends on the amplitude of synaptic inputs, synaptic plasticity is expected to affect the extent of gain modulation. In the cerebellum, synapses between mossy fibres (MFs) and granule cells (GCs) undergo short-term depression (STD) [5]. We used computer simulations and electrophysiological recordings to study the effect of STD on gain modulation by tonic inhibition in GCs. We injected depressing and non-depressing conductance trains into GCs in cerebellar slices and found that short-term depression (STD) of synaptic inputs resulted in a large enhancement of gain modulation by a tonic inhibitory conductance. STD has two effects: it reduces the amount of noise in the inputs, and it introduces a non-linear relationship between input rate and mean input conductance. Using a simple STD model, we could show that the effect of the non-linearity dominated the noise effect for all levels of STD. Our results suggest that the presence of STD at the MF-GC cell synapse could optimise gain modulation at the cerebellar input stage.