{"id":"arxiv/2605.19609","url":"https://arxiv.org/abs/2605.19609","tags":["math.OC"],"score":0,"title":"GLOBe: A Modular Global Optimization library","domain":"math.OC","source":"arxiv","authors":["Gaëtan Serré","Argyris Kalogeratos","Nicolas Vayatis"],"reviews":0,"subject":"math.OC","summary":"Open-source libraries are have a catalytic role in research pipelines, where new methods must be compared against up-to-date baselines. We present the GLobal Optimization Benchmark (GLOBe) modular Python library that unifies classical and recent continuous global optimization algorithms, including decision-based and mathematically founded particle-based methods, in a single framework. A central contribution of GLOBe is its modular architecture, which factors common algorithmic patterns into reusable familyl evel components and enables plugins to be implemented once and makes them available…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-19","reactions":{}}