Particle Data Group - 2017 Review has some strong lower limits for the mass scales of possible quark and lepton compositeness, or at least the compositeness of the easier-to-study ones, like up and down quarks and also electrons. The limits are well into the TeV range, though they are somewhat model-dependent. This means that electrons, up quarks, and down quarks do not start to disintegrate even after applying energies a million times their rest masses. Here are the maximum ratios of disintegration energy to rest mass for entities previously discovered to have been composite: Atoms: 10^(-8) (ionization of hydrogen atoms) Nuclei: 10^(-3) (dissociation of deuterons) Hadrons: 1 (deep inelastic scattering off of nucleons) That makes it very difficult for Standard-Model elementary fermions to be composite: they are much less massive than their compositeness energy scales. But there is a theoretical analogy: light mesons, like pions. Their mass is roughly sqrt(mq*mc) where mq is the quarks' average mass and mc = QCD energy scale. For mq << mc, m(meson) << mc also. So if the electron has a compositeness scale of about 1 TeV, then the electron's constituents would have to have masses around 1 eV.