Fusarium graminearum (Fgr) is a devastating filamentous fungal pathogen that causes diseases in cereals, while producing mycotoxins that are toxic for humans and animals, and render grains unusable. Low efficiency in managing Fgr poses a constant need for identifying novel control mechanisms. Evidence that fungal extracellular vesicles (EVs) from pathogenic yeast have a role in human disease led us to question whether this is also true for fungal plant pathogens. We separated EVs from Fgr and performed a proteomic analysis to determine if EVs carry proteins with potential roles in pathogenesis. We revealed that protein effectors, which are crucial for fungal virulence, were detected in EV preparations and some of them did not contain predicted secretion signals. Furthermore, a transcriptomic analysis of corn (Zea mays) plants infected by Fgr revealed that the genes of some of the effectors were highly expressed in vivo, suggesting that the Fgr EVs are a mechanism for the unconventional secretion of effectors and virulence factors. Our results expand the knowledge on fungal EVs in plant pathogenesis and cross-kingdom communication, and may contribute to the discovery of new antifungals. The following are available online at www.mdpi.com/xxx/s1, Figure S1. Controls for the separation of EVs from Fusarium graminearum (Fgr) by SEC. Figure S2. The superoxide dismutase [Cu-Zn] (SOD1) from F. graminearum (Fgr) contains a diacidic amino acid motif implicated in unconventional secretion. Figure S3. Sequence alignment of the chitinase GH18 domain. Figure S4. Computational prediction of effector candidates detected in EV samples from Fusarium graminearum (Fgr). Table S1. List of proteins detected in EVs from Fusarium graminearum (Fgr). Table S2. List of proteins employed in the computational effector prediction analysis. Table S3. Proteins identified in the secretome from Fusarium graminearum (Fgr). Table S4. List of transcripts identified in corn (Zea mays) infected by Fusarium graminearum (Fgr). Table S5. Gene expression values per biological replicate.