OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

BERTimbau

BERTimbau is a repository of pretrained, cased BERT-Base and BERT-Large models for Portuguese, with TensorFlow and PyTorch artifacts and usage materials for downstream NLP tasks.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

BERTimbau — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use BERTimbau when you need pretrained Portuguese BERT variants and want an example for loading the PyTorch artifacts or materials for reproducing the repository’s NER experiments.

Who it is for

  • Developers building Portuguese-language NLP applications.
  • Researchers exploring Portuguese text similarity, textual entailment or named-entity recognition.

Documented uses

  • Load a pretrained Portuguese model and tokenizer for a Transformers-based application.
  • Use Portuguese BERT representations as a starting point for similarity or entailment work.
  • Reproduce the repository’s named-entity recognition experiments using its experiment code and instructions.
Before you choose

Requirements & limits

  • Both BERT-Base and BERT-Large are offered, with cased variants.
  • TensorFlow and PyTorch artifacts are described; the detailed loading example uses PyTorch with Hugging Face Transformers.
  • Reported benchmark results use ASSIN2 for STS and RTE, and MiniHAREM for NER; results may not predict performance on other data.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Load a pretrained model with Hugging Face Transformers

Load the BERTimbau tokenizer and PyTorch model using the repository’s documented Transformers example.

Before you start

  • Python
  • Hugging Face Transformers
  • Choose BERT-Base or BERT-Large

Steps

  1. Import AutoModel and AutoTokenizer from transformers.

  2. Call from_pretrained with the matching model identifier: neuralmind/bert-base-portuguese-cased or neuralmind/bert-large-portuguese-cased.

  3. Use the returned tokenizer and model in your Transformers-based NLP application.

Expected outcome

A tokenizer and pretrained PyTorch model loaded through Transformers.

Guide limitations

  • The example shows model loading, not a complete task-specific application or production pipeline.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
neuralmind-ai/portuguese-bert · Official homepage title · retained in its source language
How we review sourcesSuggest a correction