The M5 muscarinic acetylcholine receptor (mAChR) is a potential therapeutic target for various neurological disorders due to its exclusive expression in dopamine-containing neurons of the brain. However, validation of the M5 mAChR as a therapeutic target is limited by a lack of M5 mAChR selective tool compounds due to the highly conserved neurotransmitter binding site at all mAChR subtypes. To circumvent this problem, drug discovery has turned to allosteric modulators that target sites that have lower amino acid conservation. Selective allosteric modulators such as ML375 and ML380 have been discovered for the M5 mAChR and these have shown promise in preclinical studies of addiction and anxiety. Whilst these novel modulators are valuable tool compounds, they are considered poor therapeutic candidates due to poor drug metabolism and pharmacokinetic profiles. Developing improved allosteric modulators is difficult as it is unknown where these allosteric modulators bind to the M5 mAChR. To aid the development of improved M5 mAChR selective allosteric modulators, we determined the allosteric binding sites of ML375 and ML380 through molecular docking, pharmacological interrogation with receptor mutants/chimeras, and cryo-electron microscopy (cryo-EM). Pharmacological results show that ML375 binds to the M5 mAChR through non-conserved residues in the upper part of an extrahelical groove between transmembrane domains 2, 3, and 4 but that ML380 does not use the same residues. Instead, a 2.8 Å cryo-EM structure of the M5 mAChR with ML380 bound revealed potential ligand density at the bottom part of the extrahelical groove. Site-directed mutagenesis experiments validated the binding site for ML380. Our work highlights the many ways allosteric ligands modulate mAChR function, and the strength of a combined pharmacological and cryo-EM approach to discover novel allosteric sites that pave the way for future drug discovery efforts at this promising target.