diff --git a/model_cards/allenai/wmt16-en-de-12-1/README.md b/model_cards/allenai/wmt16-en-de-12-1/README.md
index a79089024d5..f2009b4a1fd 100644
--- a/model_cards/allenai/wmt16-en-de-12-1/README.md
+++ b/model_cards/allenai/wmt16-en-de-12-1/README.md
@@ -1,7 +1,8 @@
 
 ---
-
-language: en, de
+language:
+- en
+- de
 thumbnail:
 tags:
 - translation
diff --git a/model_cards/allenai/wmt16-en-de-dist-12-1/README.md b/model_cards/allenai/wmt16-en-de-dist-12-1/README.md
index dd5f70da88c..d9b48afa538 100644
--- a/model_cards/allenai/wmt16-en-de-dist-12-1/README.md
+++ b/model_cards/allenai/wmt16-en-de-dist-12-1/README.md
@@ -1,7 +1,8 @@
 
 ---
-
-language: en, de
+language:
+- en
+- de
 thumbnail:
 tags:
 - translation
@@ -21,7 +22,7 @@ metrics:
 
 This is a ported version of fairseq-based [wmt16 transformer](https://github.com/jungokasai/deep-shallow/) for en-de.
 
-For more details, please see, [Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation](https://arxiv.org/abs/2006.10369).
+For more details, please, see [Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation](https://arxiv.org/abs/2006.10369).
 
 All 3 models are available:
 
diff --git a/model_cards/allenai/wmt16-en-de-dist-6-1/README.md b/model_cards/allenai/wmt16-en-de-dist-6-1/README.md
index 29c94125f53..3f37e6eaaf2 100644
--- a/model_cards/allenai/wmt16-en-de-dist-6-1/README.md
+++ b/model_cards/allenai/wmt16-en-de-dist-6-1/README.md
@@ -1,7 +1,8 @@
 
 ---
-
-language: en, de
+language:
+- en
+- de
 thumbnail:
 tags:
 - translation
diff --git a/model_cards/allenai/wmt19-de-en-6-6-base/README.md b/model_cards/allenai/wmt19-de-en-6-6-base/README.md
index 6b841b74e87..6cf44127d26 100644
--- a/model_cards/allenai/wmt19-de-en-6-6-base/README.md
+++ b/model_cards/allenai/wmt19-de-en-6-6-base/README.md
@@ -1,7 +1,9 @@
 
 ---
 
-language: de, en
+language:
+- de
+- en
 thumbnail:
 tags:
 - translation
diff --git a/model_cards/allenai/wmt19-de-en-6-6-big/README.md b/model_cards/allenai/wmt19-de-en-6-6-big/README.md
index 271a56b80e9..a5c559b0594 100644
--- a/model_cards/allenai/wmt19-de-en-6-6-big/README.md
+++ b/model_cards/allenai/wmt19-de-en-6-6-big/README.md
@@ -1,7 +1,9 @@
 
 ---
 
-language: de, en
+language:
+- de
+- en
 thumbnail:
 tags:
 - translation
diff --git a/model_cards/facebook/wmt19-de-en/README.md b/model_cards/facebook/wmt19-de-en/README.md
index 4cbb36e2e7a..e61f041a20e 100644
--- a/model_cards/facebook/wmt19-de-en/README.md
+++ b/model_cards/facebook/wmt19-de-en/README.md
@@ -1,13 +1,13 @@
 
 ---
-
-<!-- This file has been auto-generated by src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py - DO NOT EDIT or your changes will be lost -->
-
-language: de, en
+language: 
+- de
+- en
 thumbnail:
 tags:
 - translation
 - wmt19
+- facebook
 license: Apache 2.0
 datasets:
 - http://www.statmt.org/wmt19/ ([test-set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561))
diff --git a/model_cards/facebook/wmt19-en-de/README.md b/model_cards/facebook/wmt19-en-de/README.md
index 426082eaac3..1592333678e 100644
--- a/model_cards/facebook/wmt19-en-de/README.md
+++ b/model_cards/facebook/wmt19-en-de/README.md
@@ -1,13 +1,13 @@
 
 ---
-
-<!-- This file has been auto-generated by src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py - DO NOT EDIT or your changes will be lost -->
-
-language: en, de
+language: 
+- en
+- de
 thumbnail:
 tags:
 - translation
 - wmt19
+- facebook
 license: Apache 2.0
 datasets:
 - http://www.statmt.org/wmt19/ ([test-set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561))
diff --git a/model_cards/facebook/wmt19-en-ru/README.md b/model_cards/facebook/wmt19-en-ru/README.md
index 26999e65253..0cf7e1a7112 100644
--- a/model_cards/facebook/wmt19-en-ru/README.md
+++ b/model_cards/facebook/wmt19-en-ru/README.md
@@ -1,13 +1,13 @@
 
 ---
-
-<!-- This file has been auto-generated by src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py - DO NOT EDIT or your changes will be lost -->
-
-language: en, ru
+language: 
+- en
+- ru
 thumbnail:
 tags:
 - translation
 - wmt19
+- facebook
 license: Apache 2.0
 datasets:
 - http://www.statmt.org/wmt19/ ([test-set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561))
diff --git a/model_cards/facebook/wmt19-ru-en/README.md b/model_cards/facebook/wmt19-ru-en/README.md
index a4071e4101a..96c63fdaf4b 100644
--- a/model_cards/facebook/wmt19-ru-en/README.md
+++ b/model_cards/facebook/wmt19-ru-en/README.md
@@ -1,13 +1,13 @@
 
 ---
-
-<!-- This file has been auto-generated by src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py - DO NOT EDIT or your changes will be lost -->
-
-language: ru, en
+language: 
+- ru
+- en
 thumbnail:
 tags:
 - translation
 - wmt19
+- facebook
 license: Apache 2.0
 datasets:
 - http://www.statmt.org/wmt19/ ([test-set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561))
diff --git a/scripts/fsmt/gen-card-allenai-wmt16.py b/scripts/fsmt/gen-card-allenai-wmt16.py
index 854d2a11fb3..aa5a0ed3c4c 100755
--- a/scripts/fsmt/gen-card-allenai-wmt16.py
+++ b/scripts/fsmt/gen-card-allenai-wmt16.py
@@ -25,8 +25,9 @@ def write_model_card(model_card_dir, src_lang, tgt_lang, model_name):
 
     readme = f"""
 ---
-
-language: {src_lang}, {tgt_lang}
+language:
+- {src_lang}
+- {tgt_lang}
 thumbnail:
 tags:
 - translation
diff --git a/scripts/fsmt/gen-card-allenai-wmt19.py b/scripts/fsmt/gen-card-allenai-wmt19.py
index 0214e5190d3..1d7dd9964e9 100755
--- a/scripts/fsmt/gen-card-allenai-wmt19.py
+++ b/scripts/fsmt/gen-card-allenai-wmt19.py
@@ -25,7 +25,9 @@ def write_model_card(model_card_dir, src_lang, tgt_lang, model_name):
     readme = f"""
 ---
 
-language: {src_lang}, {tgt_lang}
+language:
+- {src_lang}
+- {tgt_lang}
 thumbnail:
 tags:
 - translation
diff --git a/scripts/fsmt/gen-card-facebook-wmt19.py b/scripts/fsmt/gen-card-facebook-wmt19.py
index ceef013de33..8e281f5d7cd 100755
--- a/scripts/fsmt/gen-card-facebook-wmt19.py
+++ b/scripts/fsmt/gen-card-facebook-wmt19.py
@@ -26,14 +26,14 @@ def write_model_card(model_card_dir, src_lang, tgt_lang):
 
     readme = f"""
 ---
-
-<!-- This file has been auto-generated by src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py - DO NOT EDIT or your changes will be lost -->
-
-language: {src_lang}, {tgt_lang}
+language: 
+- {src_lang}
+- {tgt_lang}
 thumbnail:
 tags:
 - translation
 - wmt19
+- facebook
 license: Apache 2.0
 datasets:
 - http://www.statmt.org/wmt19/ ([test-set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561))
@@ -69,7 +69,7 @@ mname = "facebook/wmt19-{src_lang}-{tgt_lang}"
 tokenizer = FSMTTokenizer.from_pretrained(mname)
 model = FSMTForConditionalGeneration.from_pretrained(mname)
 
-input = "{texts[src_lang]}
+input = "{texts[src_lang]}"
 input_ids = tokenizer.encode(input, return_tensors="pt")
 outputs = model.generate(input_ids)
 decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
diff --git a/src/transformers/modeling_fsmt.py b/src/transformers/modeling_fsmt.py
index a0c914cedef..7e8ddc8f9f1 100644
--- a/src/transformers/modeling_fsmt.py
+++ b/src/transformers/modeling_fsmt.py
@@ -56,8 +56,7 @@ logger = logging.get_logger(__name__)
 _CONFIG_FOR_DOC = "FSMTConfig"
 _TOKENIZER_FOR_DOC = "FSMTTokenizer"
 
-# See all FSMT models at https://huggingface.co/models?search=fsmt
-
+# See all FSMT models at https://huggingface.co/models?filter=fsmt
 
 # Porting notes:
 # this one is modeled after BartModel*