This dataset contains the processed graph dataset and trained model checkpoint for the NuGraph2 model architecture, as described in arXiv:2403.11872, compressed into a single zip archive. The graph dataset is derived from the MicroBooNE open dataset, and is stored in HDF5 file format as NuGraph2-paper.gnn.h5. A trained model architecture is also provided as NuGraph2-paper.ckpt. The repository containing the dataloader and model architecture used to interface with this dataset is available on GitHub, and can be installed via PyPi or conda. We acknowledge the MicroBooNE Collaboration for making publicly available the data sets [DOIs 10.5281/zenodo.8370883 and 10.5281/zenodo.7261921] employed in this work. These data sets consist of simulated neutrino interactions from the Booster Neutrino Beamline overlaid on top of cosmic data collected with the MicroBooNE detector [2017 JINST 12 P02017].