Existing retinal vessel segmentation datasets are mostly fundus images. We release this multi-modal retinal vessel segmentation dataset with our paper "Universal Vessel Segmentation for Multi-Modality Retinal Images" (IEEE Trans. Image Processing, vol.34, pp.7903-7918, 2025) to facilitate the study of vessel segmentation in multi-modal retinal imaging. The dataset has three subdatasets: JRCFA, JRCFAF and JRCIR, for the Fluorescence Angiography (FA), Fundus AutoFluorescence (FAF) and Infrard Reflectance (IR) modalities, respectively. Each dataset contains 40 images with high-quality vessel annotations by retina experts in the Jacobs Retina Center (JRC) in University of California, San Diego.