This technical report presents Sabi\'a-4 and Sabiazinho-4, a new generation of Portuguese language models with a focus on Brazilian Portuguese language. The models were developed through a four-stage training pipeline: continued pre-training on Portuguese and Brazilian legal corpora, long-context extension to 128K tokens, supervised fine-tuning on instruction data spanning chat, code, legal tasks, and function calling, and preference alignment. We evaluate the models on six benchmark categories: conversational capabilities in Brazilian Portuguese, knowledge of Brazilian legislation, long-context understanding, instruction following, standardized exams, and agentic capabilities including tool use and web navigation. Results show that Sabi\'a-4 and Sabiazinho-4 achieve a favorable cost-performance trade-off compared to other models, positioning them in the upper-left region of the pricing-accuracy chart. The models show improvements over previous generations in legal document drafting, multi-turn dialogue quality, and agentic task completion.
Cite
@article{arxiv.2603.10213,
title = {Sabi\'a-4 Technical Report},
author = {Thiago Laitz and Thales Sales Almeida and Hugo Abonizio and Roseval Malaquias Junior and Giovana Kerche Bonás and Marcos Piau and Celio Larcher and Ramon Pires and Rodrigo Nogueira},
journal= {arXiv preprint arXiv:2603.10213},
year = {2026}
}