diff --git a/src/transformers/models/blenderbot/modeling_blenderbot.py b/src/transformers/models/blenderbot/modeling_blenderbot.py
index 76fb22e53d9..cdaa21521c5 100755
--- a/src/transformers/models/blenderbot/modeling_blenderbot.py
+++ b/src/transformers/models/blenderbot/modeling_blenderbot.py
@@ -1119,22 +1119,22 @@ class BlenderbotModel(BlenderbotPreTrainedModel):
     @replace_return_docstrings(output_type=Seq2SeqModelOutput, config_class=_CONFIG_FOR_DOC)
     def forward(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
-        encoder_outputs=None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        decoder_input_ids: Optional[torch.LongTensor] = None,
+        decoder_attention_mask: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        decoder_head_mask: Optional[torch.Tensor] = None,
+        cross_attn_head_mask: Optional[torch.Tensor] = None,
+        encoder_outputs: Optional[Union[Tuple, BaseModelOutput]] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        decoder_inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.FloatTensor], Seq2SeqModelOutput]:
         r"""
         Returns:
 
@@ -1275,23 +1275,23 @@ class BlenderbotForConditionalGeneration(BlenderbotPreTrainedModel):
     @add_end_docstrings(BLENDERBOT_GENERATION_EXAMPLE)
     def forward(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
-        encoder_outputs=None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        labels=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        decoder_input_ids: Optional[torch.LongTensor] = None,
+        decoder_attention_mask: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        decoder_head_mask: Optional[torch.Tensor] = None,
+        cross_attn_head_mask: Optional[torch.Tensor] = None,
+        encoder_outputs: Optional[Union[Tuple, BaseModelOutput]] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        decoder_inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.FloatTensor], Seq2SeqLMOutput]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
             Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
diff --git a/src/transformers/models/blenderbot/modeling_tf_blenderbot.py b/src/transformers/models/blenderbot/modeling_tf_blenderbot.py
index e4d992cca9e..34a5f972abc 100644
--- a/src/transformers/models/blenderbot/modeling_tf_blenderbot.py
+++ b/src/transformers/models/blenderbot/modeling_tf_blenderbot.py
@@ -18,7 +18,7 @@
 import os
 import random
 import warnings
-from typing import Optional, Tuple, Union
+from typing import List, Optional, Tuple, Union
 
 import tensorflow as tf
 
@@ -1137,24 +1137,24 @@ class TFBlenderbotModel(TFBlenderbotPreTrainedModel):
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
+        input_ids: Optional[tf.Tensor] = None,
+        attention_mask: Optional[tf.Tensor] = None,
+        decoder_input_ids: Optional[tf.Tensor] = None,
+        decoder_attention_mask: Optional[tf.Tensor] = None,
+        head_mask: Optional[tf.Tensor] = None,
+        decoder_head_mask: Optional[tf.Tensor] = None,
+        cross_attn_head_mask: Optional[tf.Tensor] = None,
         encoder_outputs: Optional[Union[Tuple, TFBaseModelOutput]] = None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        training=False,
+        past_key_values: Optional[List[tf.Tensor]] = None,
+        inputs_embeds: Optional[tf.Tensor] = None,
+        decoder_inputs_embeds: Optional[tf.Tensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        training: Optional[bool] = False,
         **kwargs
-    ):
+    ) -> Union[Tuple[tf.Tensor], TFSeq2SeqModelOutput]:
         outputs = self.model(
             input_ids=input_ids,
             attention_mask=attention_mask,
@@ -1253,25 +1253,25 @@ class TFBlenderbotForConditionalGeneration(TFBlenderbotPreTrainedModel, TFCausal
     @add_end_docstrings(BLENDERBOT_GENERATION_EXAMPLE)
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
-        encoder_outputs: Optional[TFBaseModelOutput] = None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
+        input_ids: Optional[tf.Tensor] = None,
+        attention_mask: Optional[tf.Tensor] = None,
+        decoder_input_ids: Optional[tf.Tensor] = None,
+        decoder_attention_mask: Optional[tf.Tensor] = None,
+        head_mask: Optional[tf.Tensor] = None,
+        decoder_head_mask: Optional[tf.Tensor] = None,
+        cross_attn_head_mask: Optional[tf.Tensor] = None,
+        encoder_outputs: Optional[Union[Tuple, TFBaseModelOutput]] = None,
+        past_key_values: Optional[List[tf.Tensor]] = None,
+        inputs_embeds: Optional[tf.Tensor] = None,
+        decoder_inputs_embeds: Optional[tf.Tensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[tf.Tensor] = None,
+        training: Optional[bool] = False,
         **kwargs,
-    ):
+    ) -> Union[Tuple[tf.Tensor], TFSeq2SeqLMOutput]:
         r"""
         labels (`tf.tensor` of shape `(batch_size, sequence_length)`, *optional*):
             Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
diff --git a/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py b/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py
index 6c341c4cb86..8a6c863e633 100755
--- a/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py
+++ b/src/transformers/models/blenderbot_small/modeling_blenderbot_small.py
@@ -1102,22 +1102,22 @@ class BlenderbotSmallModel(BlenderbotSmallPreTrainedModel):
     @replace_return_docstrings(output_type=Seq2SeqModelOutput, config_class=_CONFIG_FOR_DOC)
     def forward(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
-        encoder_outputs=None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        decoder_input_ids: Optional[torch.LongTensor] = None,
+        decoder_attention_mask: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        decoder_head_mask: Optional[torch.Tensor] = None,
+        cross_attn_head_mask: Optional[torch.Tensor] = None,
+        encoder_outputs: Optional[Union[Tuple, BaseModelOutput]] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        decoder_inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.FloatTensor], Seq2SeqModelOutput]:
         r"""
         Returns:
 
@@ -1246,23 +1246,23 @@ class BlenderbotSmallForConditionalGeneration(BlenderbotSmallPreTrainedModel):
     @add_end_docstrings(BLENDERBOT_SMALL_GENERATION_EXAMPLE)
     def forward(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
-        encoder_outputs=None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        labels=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-    ):
+        input_ids: Optional[torch.LongTensor] = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        decoder_input_ids: Optional[torch.LongTensor] = None,
+        decoder_attention_mask: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        decoder_head_mask: Optional[torch.Tensor] = None,
+        cross_attn_head_mask: Optional[torch.Tensor] = None,
+        encoder_outputs: Optional[Union[Tuple, BaseModelOutput]] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.Tensor] = None,
+        decoder_inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.FloatTensor], Seq2SeqLMOutput]:
         r"""
         labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
             Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
diff --git a/src/transformers/models/blenderbot_small/modeling_tf_blenderbot_small.py b/src/transformers/models/blenderbot_small/modeling_tf_blenderbot_small.py
index bb2149c8acd..69d6233f025 100644
--- a/src/transformers/models/blenderbot_small/modeling_tf_blenderbot_small.py
+++ b/src/transformers/models/blenderbot_small/modeling_tf_blenderbot_small.py
@@ -16,7 +16,7 @@
 
 
 import random
-from typing import Optional, Tuple, Union
+from typing import List, Optional, Tuple, Union
 
 import numpy as np
 import tensorflow as tf
@@ -1132,24 +1132,24 @@ class TFBlenderbotSmallModel(TFBlenderbotSmallPreTrainedModel):
     )
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
+        input_ids: Optional[tf.Tensor] = None,
+        attention_mask: Optional[tf.Tensor] = None,
+        decoder_input_ids: Optional[tf.Tensor] = None,
+        decoder_attention_mask: Optional[tf.Tensor] = None,
+        head_mask: Optional[tf.Tensor] = None,
+        decoder_head_mask: Optional[tf.Tensor] = None,
+        cross_attn_head_mask: Optional[tf.Tensor] = None,
         encoder_outputs: Optional[Union[Tuple, TFBaseModelOutput]] = None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        training=False,
+        past_key_values: Optional[List[tf.Tensor]] = None,
+        inputs_embeds: Optional[tf.Tensor] = None,
+        decoder_inputs_embeds: Optional[tf.Tensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        training: Optional[bool] = False,
         **kwargs
-    ):
+    ) -> Union[Tuple[tf.Tensor], TFSeq2SeqModelOutput]:
 
         outputs = self.model(
             input_ids=input_ids,
@@ -1236,25 +1236,25 @@ class TFBlenderbotSmallForConditionalGeneration(TFBlenderbotSmallPreTrainedModel
     @add_end_docstrings(BLENDERBOT_SMALL_GENERATION_EXAMPLE)
     def call(
         self,
-        input_ids=None,
-        attention_mask=None,
-        decoder_input_ids=None,
-        decoder_attention_mask=None,
-        head_mask=None,
-        decoder_head_mask=None,
-        cross_attn_head_mask=None,
+        input_ids: Optional[tf.Tensor] = None,
+        attention_mask: Optional[tf.Tensor] = None,
+        decoder_input_ids: Optional[tf.Tensor] = None,
+        decoder_attention_mask: Optional[tf.Tensor] = None,
+        head_mask: Optional[tf.Tensor] = None,
+        decoder_head_mask: Optional[tf.Tensor] = None,
+        cross_attn_head_mask: Optional[tf.Tensor] = None,
         encoder_outputs: Optional[TFBaseModelOutput] = None,
-        past_key_values=None,
-        inputs_embeds=None,
-        decoder_inputs_embeds=None,
-        use_cache=None,
-        output_attentions=None,
-        output_hidden_states=None,
-        return_dict=None,
-        labels=None,
-        training=False,
+        past_key_values: Optional[List[tf.Tensor]] = None,
+        inputs_embeds: Optional[tf.Tensor] = None,
+        decoder_inputs_embeds: Optional[tf.Tensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+        labels: Optional[tf.Tensor] = None,
+        training: Optional[bool] = False,
         **kwargs,
-    ):
+    ) -> Union[Tuple[tf.Tensor], TFSeq2SeqLMOutput]:
         r"""
         labels (`tf.tensor` of shape `(batch_size, sequence_length)`, *optional*):
             Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,