Context. Large spectroscopic surveys of individual massive stars, such as ULLYSES and XS hoot U, provide observational data for hundreds of massive stars. Their analysis requires large numbers of synthetic spectra so that stellar parameters can be determined. In addition, libraries of massive stars’ spectra are needed to produce population synthesis models able to reproduce the observed spectra of unresolved young stellar populations, such as those revealed by the James Webb Space Telescope (JWST) in the early Universe. Aims. Our main goal is to provide an extensive library of synthetic spectra and spectral energy distributions of OB stars at metallicities of the Magellanic Clouds. This library will offer a wealth of spectrophotometric information, making it readily applicable to a variety of astrophysical problems. Methods. We used the CMFGEN code to calculate 606 NLTE, line-blanketed, expanding atmosphere models using a comprehensive set of atomic data. An overall metallicity of 1/2 Z ⊙ and 1/5 Z ⊙ was adopted for the Large Magellanic Cloud (LMC) and Small Magellanic Cloud (SMC), respectively. We produced high-resolution spectra from 30 Å to 3 µm for stars on the Main Sequence and slightly beyond. Results. We provide spectral energy distributions, normalized synthetic spectra, ionizing fluxes, and photometry in various bands: Johnson UBV, Cousins RI, Bessel JHK, selected wide JWST filters, Gaia, and LSST ugrizy filters. For each of these filters, we compute bolometric corrections for all synthetic spectra and calibrations as a function of effective temperature. Conclusions. All of our synthetic spectra are publicly available through the POLLUX database, aiming to expedite multiwavelength analyses of massive stars in low metallicity environments.