Membrane fusion occurs at the early stages of SARS-CoV-2 replication, during entry of the virus, and later during the formation of multinucleated cells called syncytia. Fusion is mediated by the binding of the viral Spike protein to its receptor ACE2. Lipid rafts are dynamic nanodomains enriched in cholesterol and sphingolipids. Rafts can act as platforms for entry of different viruses by localizing virus receptors, and attachment factors to the same membrane domains. Here, we first demonstrate that cholesterol depletion by methyl-beta-cyclodextrin inhibits Spike-mediated fusion and entry. To further study the role of ACE2 lipid raft localization in SARS-CoV-2 fusion and entry, we designed a GPI-anchored ACE2 construct. Both ACE2 and ACE2-GPI proteins were similarly expressed at the plasma membrane. Through membrane flotation assays, we show that in different cell lines, ACE2-GPI localizes predominantly to raft domains of the plasma membrane while ACE2 is non-raft associated. We then compare the ability of ACE2 and ACE2-GPI to permit SARS-CoV-2 entry, replication, and syncytia formation of different viral variants. We find little difference in the two proteins. Our results demonstrate that SARS-CoV-2 entry and fusion are cholesterol-dependent and raft-independent processes.IMPORTANCERafts are often exploited by viruses and used as platforms to enhance their entry into the cell or spread from cell to cell. The membrane localization of ACE2 and the role of lipid rafts in SARS-CoV-2 entry and cell-to-cell spread are poorly understood. The function of lipid rafts in viral fusion is often studied through their disruption by cholesterol-depleting agents. However, this process may have off-target impacts on viral fusion independently of lipid-raft disruption. Therefore, we created an ACE2 construct that localizes to lipid rafts using a GPI anchor. Conversely, wild-type ACE2 was non-raft associated. We find that the localization of ACE2 to lipid rafts does not modify the fusion dynamics of SARS-CoV-2.