Abstract Cannabinoid CB1 and CB2 receptors are members of the G protein-coupled receptor family, which is the largest class of membrane proteins in the human genome. As part of the endocannabinoid system, they have many regulatory functions in the human body. Their malfunction therefore triggers a diverse set of undesired conditions, such as pain, neuropathy, nephropathy, pruritus, osteoporosis, cachexia and Alzheimer’s disease. Although drugs targeting the system exist, the molecular and functional mechanisms involved are still poorly understood, preventing the development of better therapeutics with fewer undesired effects. One path toward the development of better and safer medicines targeting cannabinoid receptors relies on the ability of some compounds to activate a subset of pathways engaged by the receptor while sparing or even inhibiting the others, a phenomenon known as biased signaling. To take advantage of this phenomenon for drug development, a better profiling of the pathways engaged by the receptors is required. Using a BRET-based signaling detection platform, we systematically analyzed the primary signaling cascades activated by CB1 and CB2 receptors, including 9 G protein and 2 β-arrestin subtypes. Given that biased signaling is driven by ligand-specific distinct active conformations of the receptor, establishing a link between the signaling profiles elicited by different drugs and their chemotypes may help designing compounds that selectively activate beneficial pathways while avoiding those leading to undesired effects. We screened a selection of 35 structurally diverse ligands, including endocannabinoids, phytocannabinoids and synthetic compounds structurally similar or significantly different from natural cannabinoids. Our data show that biased signaling is a prominent feature of the cannabinoid receptor system and that, as predicted, ligands with different chemotypes have distinct signaling profiles. The study therefore allows for better understanding of cannabinoid receptors signaling and provides the information about tool compounds that can now be used to link signaling pathways to biological outcomes, aiding the design of improved therapeutics.