Magnetoelastic interactions couple the motion of atoms in a magnetic material with atomic magnetic moments, allowing to transfer mechanical and thermal energies between phonon and magnon subsystems. These interactions are responsible for many interesting phenomena such as Joule magnetostriction, the Wiedemann effect, the Villari effect, the Matteucci effect, anomalous thermal expansion, effects on sound velocity, and many others [1]. Precise control of magnetization through a mechanical excitation of the motion of atoms in magnetic materials, and vice versa, has enabled the development of a wide range of technological applications such as sensors and actuators. Presently, there are still many open questions about possible ways to model them at different spatial and time scales. Here, we discuss about the state-of-the-art of multiscale modelling of magnetoelastic phenomena, and current challenges on this topic. Starting at the smallest spatial scale (microscopic scale), the aim is to perform a quantum mechanical characterization of the magnetoelastic constants of the material at zero temperature. For instance, this task can be done through our recently developed software MAELAS [2] in an automated way. These intrinsic magnetic properties are used as inputs to build coarse-grained classical atomistic models that allow to simulate the material at larger spatial and time scales, including temperature and pressure effects, quasiparticle phenomenology like phonon-magnon scattering, dynamics and nonequilibrium phenomena, crystal phase transformation, grain boundaries, crystal defects, and nanostructures. Here, the coarse-grained modelling of atom and spin motion via spin-orbit coupling represents a key bottleneck. Recently, we proposed a methodology to construct such spin-lattice models capable to account for magnetoelastic properties based on the Néel model [3], finding quite encouraging preliminary results on the magnetoelastic effects on sound velocity [4] and temperature dependence of magnetoelasticity [5]. Finally, at the macroscopic scale, all this information can be used to construct reliable finite-element models that include magnetoelastic features, as well as material geometry, microstructure, etc. For instance, we successfully applied this type of model to study the influence of grain morphology and orientation on the saturation magnetostriction of polycrystalline Terfenol-D [6].