Background: Galectin-3 has been shown to play a key pathophysiological role in pulmonary associated inflammatory response and lung fibrosis in COVID-19 and is a mediator for viral adhesion. However, there is limited data about its potential role in severity and prognosis of COVID-19. This study aimed to investigate the predictive role of serum galectin-3 concentrations in the severe clinical outcomes of hospitalized COVID-19 patients: the severity of pneumonia, in-hospital mortality, and the need for intensive care unit (ICU) admission. Methods: This single-center study included 68 patients with laboratory- and radiologically-confirmed COVID-19 admitted to our emergency department. The study population was divided into patients with primary clinical outcomes (n = 32) and those without (n = 36). The need for ICU admission and/or in-hospital mortality were the primary clinical endpoints. The study group was also classified based on pneumonia severity: severe or mild/moderate. Blood samples were collected within 48 hours of admission to estimate serum galectin-3 concentrations. Results: Multivariate regression analysis showed that lower concentrations of galectin-3 and arterial oxygen saturation (SpO2) were independently associated with the primary clinical outcomes (OR = 0.951, p = 0.035; OR = 0.862, p = 0.017, respectively); increased concentrations of galectin-3 were an independent predictor of severe pneumonia (OR = 1.087, p = 0.016). In the receiver operating characteristics curve analysis, serum galectin-3 concentrations at hospital admission predicted pneumonia severity with 52.1% sensitivity and 90% specificity with a cutoff of 38.76 ng/mL. Conclusions: Circulating galectin-3 at hospital admission could be a useful biomarker for identifying COVID-19 patients at high risk for severe pneumonia.