We present a dataset of transcriptions of manuscripts and printed witnesses of the Middle Dutch texts Dietsche Catoen, Rhymed Bible or Scolastica (samples) and Karel ende Elegast. This open access dataset contains diplomatic transcriptions of these witnesses, amounting to a total of almost 29,000 verses. The data abides by the FAIR principles (Findable, Accessible, Interoperable and Reusable), is licensed under a CC-BY-SA license, and is available in multiple file formats. Due to its strictly diplomatic nature, this corpus is particularly suitable for research concerning textual stability, stemmatology and scribal profiling.