The availability of large high-quality single crystals is an important prerequisite for many studies in solid-state research. The optical floating-zone technique is an elegant method to grow such crystals, offering potential to prepare samples that may be hardly accessible with other techniques. As elaborated in this report, examples include single crystals with intentional compositional gradients, deliberate off-stoichiometry, or complex metallurgy. For the cubic chiral magnets Mn 1−x Fe x Si and Fe 1−x Co x Si, we prepared single crystals in which the composition was varied during growth from x = 0 − 0.15 and from x = 0.1 − 0.3, respectively. Such samples allowed us to efficiently study the evolution of the magnetic properties as a function of composition, as demonstrated by means of neutron scattering. For the archetypical chiral magnet MnSi and the itinerant antiferromagnet CrB 2 , we grew single crystals with varying initial manganese (0.99 to 1.04) and boron (1.95 to 2.1) content. Measurements of the low-temperature properties addressed the correlation between magnetic transition temperature and sample quality. Furthermore, we prepared single crystals of the diborides ErB 2 , MnB 2 , and VB 2 . In addition to high vapor pressures, these materials suffer from peritectic formation, potential decomposition, and high melting temperature, respectively.Copyright line will be provided by the publisher high vapor pressure. In addition, due to the direct optical access to the molten zone, the process parameters may be optimized already during growth.We have set up a preparation chain that is focused on growth of large high-quality single crystals of intermetallic compounds under ultra-high vacuum compatible conditions. Details on the technical aspects can be found in Refs. [1][2][3]. Core parts are three image furnaces that allow us to address a wide range of materials. The furnaces are a four-mirror image furnace from Crystal Systems