Solute clustering and precipitation in an Al–Cu–Mg–Ag–Si model alloy has been investigated by atom probe tomography (APT) as well as high-angle annular dark-field (HAADF) imaging and electron energy loss spectroscopy (EELS) in the scanning transmission electron microscope (STEM). Nine types of solute clusters (Cu, Ag, Mg–Cu, Mg–Ag, Mg–Cu–Si, Mg–Ag–Si, Mg–Ag–Cu, Cu–Ag–Si and MgAgCuSi) were observed by APT in both the as-quenched alloy and after ageing the alloy at 180 °C for 1 h. Three types of precipitates (Ω (AlCuMgAg), θ (Al ₂Cu) and Mg ₂Si) were observed by APT and HAADF-STEM after further ageing at 180 °C for 24 h and 100 h. We propose that MgAgCu and MgAgCuSi clusters are likely to be responsible for the formation of the Ω (AlCuMgAg) phase. Furthermore, we also suggest that the θ (Al ₂Cu) phase forms from Cu clusters and the Mg ₂Si phase forms from the decomposition of MgAgSi and MgAgCuSi clusters by losing Ag to Ω phase growth. Many early binary clusters (Mg–Cu, Mg–Ag) do not seem to undergo a significant further growth during ageing; these are more likely to be transformed into complex ternary and quaternary clusters and be subsequently consumed during the growth of large clusters/precipitates. Furthermore, it is proposed that the plate-like Ω (AlCuMgAg) precipitates evolve continuously from the MgAgCu and MgAgCuSi clusters, rather than via heterogeneous nucleation on their precursors (i.e. MgAgCu and MgAgCuSi clusters). More interestingly, even after ageing at 180 °C for 100 h, the Ω (AlCuMgAg) precipitates remain coherent with the α-Al matrix, indicating that these precipitates have a high thermal stability. This can mainly be attributed to the presence of a single Mg–Ag-rich monolayer observed at the interface between the Ω precipitate and the α-Al matrix, significantly improving the coarsening resistance of the Ω (AlCuMgAg) precipitates. Our results thus reveal links between a variety of solute clusters and the different types of precipitates in the Al–Cu–Mg–Ag–Si model alloy. Such information can in the future be used to control the precipitation by tailoring solute clustering.