Analgesic design and evaluation have been driven by the desire to create high-affinity high-selectivity mu (μ)-opioid peptide (MOP) receptor agonists. Such ligands are the mainstay of current clinical practice, and include morphine and fentanyl. Advances in this sphere have come from designing pharmacokinetic advantage, as in rapid metabolism for remifentanil. These produce analgesia, but also the adverse-effect profile that currently defines this drug class: ventilatory depression, tolerance, and abuse liability. The MOP receptor is part of a family, and there are significant functional interactions between other members of the family (delta [δ]-opioid peptide [DOP], kappa [κ]-opioid peptide [KOP], and nociceptin/orphanin FQ receptor [NOP]). Experimentally, MOP agonism and DOP antagonism produce anti-nociception (animals) with no tolerance, and low doses of MOP and NOP ligands synergise to antinociceptive advantage. In this latter context, the lack of effect of NOP agonists on ventilation is an additional advantage. Recent development has been to move towards low-selectivity multifunctional 'mixed ligands', such as cebranopadol, or ligand mixtures, such as Targinact®. Moreover, the observation that β-arrestin coupling underlies the side-effect profile for MOP ligands (from knockout animal studies) led to the discovery of biased (to G-protein and away from β-arrestin intracellular signalling) MOP ligands, such as oliceridine. There is sufficient excitement in the opioid field to suggest that opioid analgesics without significant side-effects may be on the horizon, and the 'opioid Holy Grail' might be in reach.