Natural regeneration is crucial for silvicultural approaches based on the continuous presence of a forest cover, or Continuous Cover Forestry (CCF). Light is considered one of the most important factors affecting regeneration growth under canopy cover. Sitka spruce, western hemlock and Douglas fir are important forestry species both in Europe and in North America with potential to be used together under CCF management. Our aim was to develop predictive early-growth models for these species growing beneath forest canopies, and to investigate species differences in terms of shade tolerance. We sampled regenerating trees growing under canopy cover at multiple sites in the UK. We compared alternative asymptotic non-linear models as a function of light availability to simulate the height growth for all species and the diameter growth for Sitka spruce and western hemlock only. We included tree size and intra-regeneration competition as predictors, which affected the asymptotic growth at full light and/or the growth rate at which such asymptote was reached. We also calibrated models of apical dominance ratio (ADR, for all species) and live crown ratio (LCR, for Douglas fir and Sitka spruce only) as a function of light availability, tree size and intra-regeneration competition. Species-specific non-linear models best simulated the light-growth responses (3-points Logistic for Sitka spruce, Michaelis-Menten for western hemlock, asymptotic with offset for Douglas fir). Tree size in all cases increased the asymptotic growth and in two cases also the growth rate. Competition significantly reduced the growth for Sitka spruce and western hemlock, with the diameter growth reduced more than height growth. Both the ADR and the LCR increased with light availability, with species-specific differences for LCR but not for ADR. For Sitka spruce an apical dominance ratio of 1.5 can be used in the field to identify adequate growing conditions. An increasing shade tolerance ranking was found as Douglas fir ≤ Sitka spruce < western hemlock. We conclude that modelling light-growth requires species-specific non-linear functions and that predictions are improved by including size and competition. The developed predictive models for height and diameter growth will allow accurate modelling of the study species in CCF management.