diff --git a/src/peft/peft_model.py b/src/peft/peft_model.py
index cb9d3bf159..c2d0275afe 100644
--- a/src/peft/peft_model.py
+++ b/src/peft/peft_model.py
@@ -31,6 +31,7 @@
 from transformers.modeling_outputs import SequenceClassifierOutput, TokenClassifierOutput
 from transformers.utils import PushToHubMixin
 
+from . import __version__
 from .tuners import (
     AdaLoraModel,
     AdaptionPromptModel,
@@ -49,7 +50,7 @@
     TaskType,
     _set_adapter,
     _set_trainable,
-    add_or_edit_model_card,
+    add_library_to_model_card,
     get_peft_model_state_dict,
     hub_file_exists,
     set_peft_model_state_dict,
@@ -128,7 +129,7 @@ def save_pretrained(self, save_directory, safe_serialization=False, **kwargs):
         if os.path.isfile(save_directory):
             raise ValueError(f"Provided path ({save_directory}) should be a directory, not a file")
         os.makedirs(save_directory, exist_ok=True)
-        add_or_edit_model_card(save_directory)
+        self.create_or_update_model_card(save_directory)
 
         for adapter_name, peft_config in self.peft_config.items():
             # save only the trainable weights
@@ -526,6 +527,46 @@ def set_adapter(self, adapter_name):
     def active_peft_config(self):
         return self.peft_config[self.active_adapter]
 
+    def create_or_update_model_card(self, output_dir):
+        """
+        Updates or create model card to include information about peft:
+        1. Adds `peft` library tag
+        2. Adds peft version
+        3. Adds quantization information if it was used
+        """
+        # Adds `peft` library tag
+        add_library_to_model_card(output_dir)
+
+        with open(os.path.join(output_dir, "README.md"), "r") as f:
+            lines = f.readlines()
+
+        quantization_config = None
+        if hasattr(self.config, "quantization_config"):
+            quantization_config = self.config.quantization_config.to_dict()
+        training_config_text = ""
+        # Adds quantization information if it was used
+        if quantization_config is not None:
+            training_config_text += "\nThe following `bitsandbytes` quantization config was used during training:\n"
+            training_config_text += "\n".join([f"- {name}: {value}" for name, value in quantization_config.items()])
+            training_config_text += "\n"
+
+        training_procedure_heading = "## Training procedure\n"
+        if training_procedure_heading in lines:
+            lines.insert(lines.index(training_procedure_heading) + 2, training_config_text)
+        else:
+            lines.append(f"{training_procedure_heading}\n{training_config_text}")
+
+        # Adds peft version
+        framework_block_heading = "### Framework versions\n"
+        if framework_block_heading in lines:
+            lines.insert(lines.index(framework_block_heading) + 2, f"- PEFT {__version__}\n")
+        else:
+            lines.append(f"{framework_block_heading}\n\n- PEFT {__version__}\n")
+
+        # write the lines back to README.md
+        with open(os.path.join(output_dir, "README.md"), "w") as f:
+            f.writelines(lines)
+
 
 class PeftModelForSequenceClassification(PeftModel):
     """
diff --git a/src/peft/utils/__init__.py b/src/peft/utils/__init__.py
index 90a93859c7..50e5da620b 100644
--- a/src/peft/utils/__init__.py
+++ b/src/peft/utils/__init__.py
@@ -27,7 +27,7 @@
     WEIGHTS_NAME,
     SAFETENSORS_WEIGHTS_NAME,
     _set_trainable,
-    add_or_edit_model_card,
+    add_library_to_model_card,
     bloom_model_postprocess_past_key_value,
     prepare_model_for_int8_training,
     prepare_model_for_kbit_training,
diff --git a/src/peft/utils/other.py b/src/peft/utils/other.py
index dc531e7d9e..971689dafb 100644
--- a/src/peft/utils/other.py
+++ b/src/peft/utils/other.py
@@ -21,7 +21,7 @@
 
 
 # Add or edit model card to have `library_name: peft`
-def add_or_edit_model_card(output_dir):
+def add_library_to_model_card(output_dir):
     if os.path.exists(os.path.join(output_dir, "README.md")):
         with open(os.path.join(output_dir, "README.md"), "r") as f:
             lines = f.readlines()