diff --git a/model_cards/facebook/bart-large-mnli/README.md b/model_cards/facebook/bart-large-mnli/README.md
index e94693864eb..1e5425ea54d 100644
--- a/model_cards/facebook/bart-large-mnli/README.md
+++ b/model_cards/facebook/bart-large-mnli/README.md
@@ -2,4 +2,7 @@
 license: mit
 thumbnail: https://huggingface.co/front/thumbnails/facebook.png
 pipeline_tag: zero-shot-classification
+widget:
+- text: "November 3 is weeks away and Biden's lead is only growing."
+  labels: "politics, economics, public health, elections"
 ---
diff --git a/model_cards/joeddav/bart-large-mnli-yahoo-answers/README.md b/model_cards/joeddav/bart-large-mnli-yahoo-answers/README.md
index 4918cc7cb1f..3ec1be34320 100644
--- a/model_cards/joeddav/bart-large-mnli-yahoo-answers/README.md
+++ b/model_cards/joeddav/bart-large-mnli-yahoo-answers/README.md
@@ -5,8 +5,7 @@ tags:
 - pytorch
 datasets:
 - yahoo-answers
-widget:
-- text: "Who are you voting for in 2020? <sep> This text is about politics."
+pipeline_tag: zero-shot-classification
 ---
 
 # bart-lage-mnli-yahoo-answers
diff --git a/model_cards/joeddav/xlm-roberta-large-xnli/README.md b/model_cards/joeddav/xlm-roberta-large-xnli/README.md
index 8f6470e4213..27ebe385867 100644
--- a/model_cards/joeddav/xlm-roberta-large-xnli/README.md
+++ b/model_cards/joeddav/xlm-roberta-large-xnli/README.md
@@ -5,11 +5,17 @@ tags:
 - pytorch
 - tensorflow
 datasets:
-- mnli
+- multi_nli
 - xnli
-widget:
-- text: "За кого вы голосуете в 2020 году? <sep> This text is about politique."
 license: mit
+pipeline_tag: zero-shot-classification
+widget:
+- text: "За кого вы голосуете в 2020 году?"
+  labels: "politique étrangère, Europe, élections, affaires, politique"
+- text: "لمن تصوت في 2020؟"
+  labels: "السياسة الخارجية, أوروبا, الانتخابات, الأعمال, السياسة"
+- text: "2020'de kime oy vereceksiniz?"
+  labels: "dış politika, Avrupa, seçimler, ticaret, siyaset"
 ---
 
 # xlm-roberta-large-xnli
@@ -115,4 +121,3 @@ This model was pre-trained on set of 100 languages, as described in
 MNLI train set and the XNLI validation and test sets. Finally, it was trained for one additional epoch on only XNLI
 data where the translations for the premise and hypothesis are shuffled such that the premise and hypothesis for
 each example come from the same original English example but the premise and hypothesis are of different languages.
-
diff --git a/model_cards/valhalla/distilbart-mnli-12-1/README.md b/model_cards/valhalla/distilbart-mnli-12-1/README.md
index 678df0a2003..fd41fa7191d 100644
--- a/model_cards/valhalla/distilbart-mnli-12-1/README.md
+++ b/model_cards/valhalla/distilbart-mnli-12-1/README.md
@@ -4,6 +4,7 @@ datasets:
 tags:
 - distilbart
 - distilbart-mnli
+pipeline_tag: zero-shot-classification
 ---
 
 # DistilBart-MNLI
diff --git a/model_cards/valhalla/distilbart-mnli-12-3/README.md b/model_cards/valhalla/distilbart-mnli-12-3/README.md
index 678df0a2003..fd41fa7191d 100644
--- a/model_cards/valhalla/distilbart-mnli-12-3/README.md
+++ b/model_cards/valhalla/distilbart-mnli-12-3/README.md
@@ -4,6 +4,7 @@ datasets:
 tags:
 - distilbart
 - distilbart-mnli
+pipeline_tag: zero-shot-classification
 ---
 
 # DistilBart-MNLI
diff --git a/model_cards/valhalla/distilbart-mnli-12-6/README.md b/model_cards/valhalla/distilbart-mnli-12-6/README.md
index 678df0a2003..fd41fa7191d 100644
--- a/model_cards/valhalla/distilbart-mnli-12-6/README.md
+++ b/model_cards/valhalla/distilbart-mnli-12-6/README.md
@@ -4,6 +4,7 @@ datasets:
 tags:
 - distilbart
 - distilbart-mnli
+pipeline_tag: zero-shot-classification
 ---
 
 # DistilBart-MNLI
diff --git a/model_cards/valhalla/distilbart-mnli-12-9/README.md b/model_cards/valhalla/distilbart-mnli-12-9/README.md
index 678df0a2003..fd41fa7191d 100644
--- a/model_cards/valhalla/distilbart-mnli-12-9/README.md
+++ b/model_cards/valhalla/distilbart-mnli-12-9/README.md
@@ -4,6 +4,7 @@ datasets:
 tags:
 - distilbart
 - distilbart-mnli
+pipeline_tag: zero-shot-classification
 ---
 
 # DistilBart-MNLI