A

Context. The James Webb Space Telescope (JWST) has uncovered an abundance of z > 10 galaxies bright in the ultraviolet (UV), whose presence poses a challenge to traditional theoretical models at high redshifts. Various new models have recently emerged to address this discrepancy by refining their description of star formation. Aims. Here, we investigate whether modifications to the stellar initial mass function (IMF) alone can reproduce the z > 10 UV luminosity functions (UV LFs) when the star formation rate is used as a proxy for the fraction of massive stars. Methods. We incorporated an evolving IMF into the <ns0:sc>ASTRAEUS</ns0:sc> galaxy evolution and reionisation simulation framework, which becomes increasingly top-heavy as the gas density in a galaxy rises above a given threshold. Our implementation accounts for the IMF’s effects on supernova (SN) feedback and metal enrichment, as well as UV and ionising emissivities. Results. For this evolving IMF model, we find that (i) the maximum UV luminosity enhancement is twice as large in massive galaxies (ΔMUV ≃ 2.6) than those where star formation is strongly limited by SN feedback (ΔMUV ≃ 1.3); (ii) it successfully reproduces the observed UV LFs at z = 5 − 15; (iii) galaxies with top-heavy IMFs exhibit the highest star formation rates, driven by their location in local density peaks, which facilitates higher gas accretion rates; (iv) the 1σ variances in the UV luminosity are only slightly higher compared to when assuming a Salpeter IMF, but the 2σ variances are significantly increased by a factor of 1.4 − 2 boosting the abundance of UV-bright galaxies at z > 10; and (v) reionisation begins earlier with more extended large ionised regions and fewer smaller ones during its initial stages, although these differences diminish at lower redshifts, leading to a similar end of reionisation at z ≃ 5.6.