Sabiá: Portuguese Large Language Models
Ramon Pires, Hugo Abonizio, Thales Sales Almeida and Rodrigo Nogueira submitted Sabiá: Portuguese Large Language Models to arXiv on April 16, 2023. The paper documents language models produced by further pretraining GPT-J and LLaMA models on Portuguese texts and evaluates them on Portuguese-language datasets.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented Portuguese-language-model research milestone focused on language-specific continued pretraining and evaluation.
People, organizations & technologies
PEOPLE
Ramon Pires, Hugo Abonizio, Thales Sales Almeida, Rodrigo Nogueira
ORGANIZATIONS
Maritaca AI
TECHNOLOGIES
Sabiá, GPT-J, LLaMA
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-10T22:06:00+08:00
Title, authors, April 16 2023 v1 submission date, Portuguese-text continued-pretraining method and Portuguese evaluation claims are supported by the original arXiv record. The record does not infer a model birth date, development geography, ancestry, influence or causality beyond what the paper explicitly states.
Sources
Sabiá: Portuguese Large Language Models
arXiv
PUBLISHED
2023-04-16
Trace the chronology
Record identity
LINEAiGE IDsabia-portuguese-llms-2023