Despite rapid progress in open large language models (LLMs), European Portuguese (pt-PT) remains underrepresented in both training data and native evaluation, with machine-translated benchmarks likely missing the variant's linguistic and cultural nuances. We introduce AMALIA, a fully open LLM that prioritizes pt-PT by using more high-quality pt-PT data during both the mid- and post-training stages. To evaluate pt-PT more faithfully, we release a suite of pt-PT benchmarks that includes translated standard tasks and four new datasets targeting pt-PT generation, linguistic competence, and pt-PT/pt-BR bias. Experiments show that AMALIA matches strong baselines on translated benchmarks while substantially improving performance on pt-PT-specific evaluations, supporting the case for targeted training and native benchmarking for European Portuguese.
@article{arxiv.2603.26511,
title = {AMALIA Technical Report: A Fully Open Source Large Language Model for European Portuguese},
author = {Afonso Simplício and Gonçalo Vinagre and Miguel Moura Ramos and Diogo Tavares and Rafael Ferreira and Giuseppe Attanasio and Duarte M. Alves and Inês Calvo and Inês Vieira and Rui Guerra and James Furtado and Beatriz Canaverde and Iago Paulo and Vasco Ramos and Diogo Glória-Silva and Miguel Faria and Marcos Treviso and Daniel Gomes and Pedro Gomes and David Semedo and André Martins and João Magalhães},
journal= {arXiv preprint arXiv:2603.26511},
year = {2026}
}
Comments
PROPOR 2026 - The 17th International Conference on Computational Processing of Portuguese