Islet inflammation and cytokine production are implicated in pancreatic beta-cell dysfunction and diabetes pathogenesis. However, we lack therapeutics to protect the insulin-producing beta-cells from inflammatory damage. Closing this clinical gap requires the establishment of new disease models of islet inflammation to facilitate screening efforts aimed at identifying new protective agents. Here, we have developed a genetic model of Interleukin-1 beta (Il-1 beta)-driven islet inflammation in zebrafish, a vertebrate that allows for non-invasive imaging of beta-cells and in vivo drug discovery. Live imaging of immune cells and beta-cells in our model revealed dynamic migration, increased visitation and prolonged macrophage retention in the islet, together with robust activation of NF-kappa B signalling in beta-cells. We find that Il-1 beta-mediated inflammation does not cause beta-cell destruction but, rather, it impairs beta-cell function and identity. In vivo, beta-cells exhibit impaired glucose-stimulated calcium influx and reduced expression of genes involved in function and maturity. These defects are accompanied by alpha-cell expansion, glucose intolerance and hyperglycemia following a glucose challenge. Notably, we show that a medicinal plant derivative (wedelolactone) is capable of reducing the immune-cell infiltration while also ameliorating the hyperglycemic phenotype of our model. Importantly, these anti-diabetic properties in zebrafish are predictive of wedelolactone's efficacy in protecting rodent and human islets fromcytokine-induced apoptosis. Insummary, this newzebrafish model of diabetes opens a window to study the interactions between immune and beta-cells in vivo, while also allowing the identification of therapeutic agents for protecting beta-cells from inflammation.