Newtrinos.jl — An Open-Source, Community-Driven Global Fit for Neutrino Physics

Global fits — combined analyses of datasets from multiple experiments — have long been central to neutrino physics, given the field’s rich and diverse experimental landscape. To date, these efforts have been carried out by a small number of groups, each relying on closed-source, proprietary tools. This approach suffers from fundamental limitations: results cannot be independently verified, code cannot be reused or built upon, and the frameworks do not scale to accommodate the growing volume and complexity of available data. Indeed, existing efforts are already struggling to incorporate certain datasets.

With critical questions in the field — neutrino mass ordering and CP violation — likely to be settled by global fits in the coming years, the stakes for getting this right have never been higher. We therefore present Newtrinos.jl, a global fit framework built from the ground up as an open-source project, and open to community contributions. Written in Julia — a modern, high-performance language designed for scientific computing — Newtrinos.jl is built for scalability and long-term maintainability. It supports modern computational workflows, including full automatic differentiation, positioning it as a robust foundation for neutrino global fitting over the next decade.