Sabiá-3 Technical Report
This provides a domain-specialized solution for users needing efficient Portuguese language processing, though it is incremental as it builds on previous models like Sabia-2 Medium.
The authors tackled the problem of developing cost-effective language models for Portuguese and Brazil-related tasks by introducing Sabiá-3 and Sabiazinho-3, which show strong performance on professional and academic benchmarks and match frontier LLMs at three to four times lower cost per token.
This report presents Sabiá-3, our new flagship language model, and Sabiazinho-3, a more cost-effective sibling. The models were trained on a large brazilian-centric corpus. Evaluations across diverse professional and academic benchmarks show a strong performance on Portuguese and Brazil-related tasks. Sabiá-3 shows large improvements in comparison to our previous best of model, Sabia-2 Medium, especially in reasoning-intensive tasks. Notably, Sabiá-3's average performance matches frontier LLMs, while it is offered at a three to four times lower cost per token, reinforcing the benefits of domain specialization.