diff --git a/model_cards/Geotrend/bert-base-15lang-cased/README.md b/model_cards/Geotrend/bert-base-15lang-cased/README.md
new file mode 100644
index 000000000000..7062aefed466
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-15lang-cased/README.md
@@ -0,0 +1,48 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-15lang-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+The measurements below have been computed on a [Google Cloud n1-standard-1 machine (1 vCPU, 3.75 GB)](https://cloud.google.com/compute/docs/machine-types\#n1_machine_type):
+|             Model               | Num parameters |   Size   |  Memory  | Loading time |
+| ------------------------------- | -------------- | -------- | -------- | ------------ |
+| bert-base-multilingual-cased    |   178 million  |  714 MB  | 1400 MB  |    4.2 sec   |
+| Geotrend/bert-base-15lang-cased |   141 million  |  564 MB  | 1098 MB  |    3.1 sec   |
+
+Handled languages: en, fr, es, de, zh, ar, ru, vi, el, bg, th, tr, hi, ur and sw.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-15lang-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-15lang-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-ar-cased/README.md b/model_cards/Geotrend/bert-base-ar-cased/README.md
new file mode 100644
index 000000000000..d8051a914ce0
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-ar-cased/README.md
@@ -0,0 +1,41 @@
+---
+language: ar
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-ar-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-ar-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-ar-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-bg-cased/README.md b/model_cards/Geotrend/bert-base-bg-cased/README.md
new file mode 100644
index 000000000000..bace35f47737
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-bg-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: bg
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-bg-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-bg-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-bg-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-de-cased/README.md b/model_cards/Geotrend/bert-base-de-cased/README.md
new file mode 100644
index 000000000000..a62a661f942a
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-de-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: de
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-de-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-de-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-de-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-el-cased/README.md b/model_cards/Geotrend/bert-base-el-cased/README.md
new file mode 100644
index 000000000000..6a0be9c55fb3
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-el-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: el
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-el-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-el-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-el-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-ar-cased/README.md b/model_cards/Geotrend/bert-base-en-ar-cased/README.md
new file mode 100644
index 000000000000..41944faa42c1
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-ar-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-ar-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-ar-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-ar-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-bg-cased/README.md b/model_cards/Geotrend/bert-base-en-bg-cased/README.md
new file mode 100644
index 000000000000..9ac9456efb58
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-bg-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-bg-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-bg-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-bg-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-cased/README.md b/model_cards/Geotrend/bert-base-en-cased/README.md
new file mode 100644
index 000000000000..c17bbf5dd5e7
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: en
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-de-cased/README.md b/model_cards/Geotrend/bert-base-en-de-cased/README.md
new file mode 100644
index 000000000000..353b24c8f9bf
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-de-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-de-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-de-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-de-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-el-cased/README.md b/model_cards/Geotrend/bert-base-en-el-cased/README.md
new file mode 100644
index 000000000000..6b8aad0f9595
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-el-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-el-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-el-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-el-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-es-cased/README.md b/model_cards/Geotrend/bert-base-en-es-cased/README.md
new file mode 100644
index 000000000000..aff383a2954b
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-es-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-es-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-es-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-es-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-fr-cased/README.md b/model_cards/Geotrend/bert-base-en-fr-cased/README.md
new file mode 100644
index 000000000000..c61ca7d078c4
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-fr-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-fr-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-fr-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-fr-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-hi-cased/README.md b/model_cards/Geotrend/bert-base-en-hi-cased/README.md
new file mode 100644
index 000000000000..0fc362533a76
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-hi-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-hi-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-hi-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-hi-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-ru-cased/README.md b/model_cards/Geotrend/bert-base-en-ru-cased/README.md
new file mode 100644
index 000000000000..98794bda1863
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-ru-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-ru-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-ru-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-ru-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-sw-cased/README.md b/model_cards/Geotrend/bert-base-en-sw-cased/README.md
new file mode 100644
index 000000000000..bf5dc89df583
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-sw-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-sw-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-sw-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-sw-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-th-cased/README.md b/model_cards/Geotrend/bert-base-en-th-cased/README.md
new file mode 100644
index 000000000000..c4e6db5a86ce
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-th-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-th-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-th-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-th-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-tr-cased/README.md b/model_cards/Geotrend/bert-base-en-tr-cased/README.md
new file mode 100644
index 000000000000..6faceb7dbcb2
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-tr-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-tr-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-tr-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-tr-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-ur-cased/README.md b/model_cards/Geotrend/bert-base-en-ur-cased/README.md
new file mode 100644
index 000000000000..ff7c258fddcd
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-ur-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-ur-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-ur-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-ur-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-vi-cased/README.md b/model_cards/Geotrend/bert-base-en-vi-cased/README.md
new file mode 100644
index 000000000000..90f34e57b1f4
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-vi-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-vi-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-vi-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-vi-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-en-zh-cased/README.md b/model_cards/Geotrend/bert-base-en-zh-cased/README.md
new file mode 100644
index 000000000000..1972efdab269
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-en-zh-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: multilingual
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-en-zh-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-en-zh-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-en-zh-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-es-cased/README.md b/model_cards/Geotrend/bert-base-es-cased/README.md
new file mode 100644
index 000000000000..7b6923412373
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-es-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: es
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-es-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-es-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-es-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-fr-cased/README.md b/model_cards/Geotrend/bert-base-fr-cased/README.md
new file mode 100644
index 000000000000..1862e1af0d02
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-fr-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: fr
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-fr-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-fr-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-fr-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-hi-cased/README.md b/model_cards/Geotrend/bert-base-hi-cased/README.md
new file mode 100644
index 000000000000..ad296021e72f
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-hi-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: hi
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-hi-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-hi-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-hi-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-ru-cased/README.md b/model_cards/Geotrend/bert-base-ru-cased/README.md
new file mode 100644
index 000000000000..1407ec4dbe67
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-ru-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: ru
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-ru-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-ru-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-ru-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-sw-cased/README.md b/model_cards/Geotrend/bert-base-sw-cased/README.md
new file mode 100644
index 000000000000..ad7ed8da6579
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-sw-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: sw
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-sw-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-sw-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-sw-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-th-cased/README.md b/model_cards/Geotrend/bert-base-th-cased/README.md
new file mode 100644
index 000000000000..b66d5e4dd263
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-th-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: th
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-th-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-th-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-th-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-tr-cased/README.md b/model_cards/Geotrend/bert-base-tr-cased/README.md
new file mode 100644
index 000000000000..a15033b96c7b
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-tr-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: tr
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-tr-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-tr-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-tr-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-ur-cased/README.md b/model_cards/Geotrend/bert-base-ur-cased/README.md
new file mode 100644
index 000000000000..931d7734a951
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-ur-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: ur
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-ur-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-ur-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-ur-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-vi-cased/README.md b/model_cards/Geotrend/bert-base-vi-cased/README.md
new file mode 100644
index 000000000000..68e4dd480094
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-vi-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: vi
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-vi-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-vi-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-vi-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.
diff --git a/model_cards/Geotrend/bert-base-zh-cased/README.md b/model_cards/Geotrend/bert-base-zh-cased/README.md
new file mode 100644
index 000000000000..8c2c947c8093
--- /dev/null
+++ b/model_cards/Geotrend/bert-base-zh-cased/README.md
@@ -0,0 +1,40 @@
+---
+language: zh
+
+datasets: wikipedia
+
+license: apache-2.0
+---
+
+# bert-base-zh-cased
+
+We are sharing smaller versions of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) that handle a custom number of languages.
+
+Unlike [distilbert-base-multilingual-cased](https://huggingface.co/distilbert-base-multilingual-cased), our versions give exactly the same representations produced by the original model which preserves the original accuracy.
+
+For more information please visit our paper: [Load What You Need: Smaller Versions of Multilingual BERT](https://www.aclweb.org/anthology/2020.sustainlp-1.16.pdf).
+
+## How to use
+
+```python
+from transformers import AutoTokenizer, AutoModel
+
+tokenizer = AutoTokenizer.from_pretrained("Geotrend/bert-base-zh-cased")
+model = AutoModel.from_pretrained("Geotrend/bert-base-zh-cased")
+
+```
+
+### How to cite
+
+```bibtex
+@inproceedings{smallermbert,
+  title={Load What You Need: Smaller Versions of Mutlilingual BERT},
+  author={Abdaoui, Amine and Pradel, Camille and Sigel, Grégoire},
+  booktitle={SustaiNLP / EMNLP},
+  year={2020}
+}
+```
+
+## Contact 
+
+Please contact amine@geotrend.fr for any question, feedback or request.