diff --git a/Options.md b/Options.md
index 569917c..f396845 100644
--- a/Options.md
+++ b/Options.md
@@ -34,6 +34,8 @@ model = BertClassifier(bert_model='bert-base-uncased',
  * `'bert-base-multilingual-uncased'`
  * `'bert-base-multilingual-cased'`
  * `'bert-base-chinese'`
+ * `'bert-base-portuguese-cased'`
+ * `'bert-large-portuguese-cased'`
  
  **SciBERT pretrained BERT models**  See [`SciBERT` github](https://github.com/allenai/scibert) and [paper](https://arxiv.org/pdf/1903.10676.pdf) for more info.
  * `'scibert-scivocab-uncased'`
diff --git a/bert_sklearn/model/pytorch_pretrained/modeling.py b/bert_sklearn/model/pytorch_pretrained/modeling.py
index ddf6d82..99b75ed 100644
--- a/bert_sklearn/model/pytorch_pretrained/modeling.py
+++ b/bert_sklearn/model/pytorch_pretrained/modeling.py
@@ -49,7 +49,9 @@
     'bert-large-cased-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-pytorch_model.bin",
     'bert-large-uncased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-pytorch_model.bin",
     'bert-large-cased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-finetuned-squad-pytorch_model.bin",
-    'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-pytorch_model.bin",    
+    'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-pytorch_model.bin",
+    'bert-base-portuguese-cased': "https://huggingface.co/neuralmind/bert-base-portuguese-cased/resolve/main/pytorch_model.bin",
+    'bert-large-portuguese-cased': "https://huggingface.co/neuralmind/bert-large-portuguese-cased/resolve/main/pytorch_model.bin",
     # BioBERT models
     'biobert-base-cased': "https://github.com/naver/biobert-pretrained/releases/download/v1.1-pubmed/biobert_v1.1_pubmed.tar.gz",
     'biobert-v1.1-pubmed-base-cased': "https://github.com/naver/biobert-pretrained/releases/download/v1.1-pubmed/biobert_v1.1_pubmed.tar.gz",
@@ -78,6 +80,8 @@
     'bert-large-uncased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-config.json",
     'bert-large-cased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-finetuned-squad-config.json",
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-config.json",
+    'bert-base-portuguese-cased': "https://huggingface.co/neuralmind/bert-base-portuguese-cased/raw/main/config.json",
+    'bert-large-portuguese-cased': "https://huggingface.co/neuralmind/bert-large-portuguese-cased/raw/main/config.json",
 }
 BERT_CONFIG_NAME = 'bert_config.json'
 TF_WEIGHTS_NAME = 'model.ckpt'
diff --git a/bert_sklearn/model/pytorch_pretrained/tokenization.py b/bert_sklearn/model/pytorch_pretrained/tokenization.py
index fbe6ecf..3db9cbc 100644
--- a/bert_sklearn/model/pytorch_pretrained/tokenization.py
+++ b/bert_sklearn/model/pytorch_pretrained/tokenization.py
@@ -42,6 +42,8 @@
     'bert-large-uncased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-uncased-whole-word-masking-finetuned-squad-vocab.txt",
     'bert-large-cased-whole-word-masking-finetuned-squad': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-large-cased-whole-word-masking-finetuned-squad-vocab.txt",
     'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-vocab.txt",
+    'bert-base-portuguese-cased': "https://huggingface.co/neuralmind/bert-base-portuguese-cased/raw/main/vocab.txt",
+    'bert-large-portuguese-cased': "https://huggingface.co/neuralmind/bert-large-portuguese-cased/raw/main/vocab.txt",
     # BioBERT models    
     'biobert-base-cased': "https://github.com/naver/biobert-pretrained/releases/download/v1.1-pubmed/biobert_v1.1_pubmed.tar.gz",
     'biobert-v1.1-pubmed-base-cased': "https://github.com/naver/biobert-pretrained/releases/download/v1.1-pubmed/biobert_v1.1_pubmed.tar.gz", 
diff --git a/demo.ipynb b/demo.ipynb
index d512026..bac9de1 100644
--- a/demo.ipynb
+++ b/demo.ipynb
@@ -32,9 +32,9 @@
     "\n",
     "* **`bert_model`** - BERT models come in 2 sizes : `base` and `large`. As you would expect the large model demands more GPU memory and takes longer to train. If you have a small GPU, start with the any of the `base` models first. The default is set to `'bert-base-uncased'`\n",
     "\n",
-    "> `base(110M parameter models)` : `'bert-base-uncased'`, `'bert-base-cased'`, `'bert-base-multilingual-uncased'`, `'bert-base-multilingual-cased'`, `'bert-base-chinese'`, and all the **`BioBERT`** and **`SciBERT`** models.\n",
+    "> `base(110M parameter models)` : `'bert-base-uncased'`, `'bert-base-cased'`, `'bert-base-multilingual-uncased'`, `'bert-base-multilingual-cased'`, `'bert-base-chinese'`, `'bert-base-portuguese-cased'`, and all the **`BioBERT`** and **`SciBERT`** models.\n",
     "\n",
-    "> `large(340M parameter models)`: `'bert-large-uncased'` and `'bert-large-cased'`\n",
+    "> `large(340M parameter models)`: `'bert-large-uncased'`, `'bert-large-cased'` and `'bert-large-portuguese-cased'`\n",
     "\n",
     "\n",
     "* **`max_seq_length`** - the defualt is 128 with a max value of 512. But seting it to a smaller value like 96 or even 64  saves a lot of GPU memory and still gets good results on a lot of tasks.\n",
diff --git a/tests/run_classifier.py b/tests/run_classifier.py
index 3943d9b..b609b41 100644
--- a/tests/run_classifier.py
+++ b/tests/run_classifier.py
@@ -71,7 +71,7 @@ def main():
     parser.add_argument("--bert_model", default=None, type=str, required=True,
                         help="Bert pre-trained model selected in the list: bert-base-uncased, "
                         "bert-large-uncased, bert-base-cased, bert-large-cased, bert-base-multilingual-uncased, "
-                        "bert-base-multilingual-cased, bert-base-chinese.")
+                        "bert-base-multilingual-cased, bert-base-chinese, bert-base-portuguese-cased, bert-large-portuguese-cased")
     parser.add_argument("--task_name",
                         default=None,
                         type=str,