diff --git a/google/generativeai/generative_models.py b/google/generativeai/generative_models.py
index 8d331a9f6..f25d0eb63 100644
--- a/google/generativeai/generative_models.py
+++ b/google/generativeai/generative_models.py
@@ -244,6 +244,7 @@ def generate_content(
         tools: content_types.FunctionLibraryType | None = None,
         tool_config: content_types.ToolConfigType | None = None,
         request_options: helper_types.RequestOptionsType | None = None,
+        extra_headers: dict[str, str] | None = None,
     ) -> generation_types.GenerateContentResponse:
         """A multipurpose function to generate responses from the model.
 
@@ -319,6 +320,14 @@ def generate_content(
         if request_options is None:
             request_options = {}
 
+        # Convert `extra_headers` to metadata format
+        if extra_headers:
+            metadata = [(k, v) for k, v in extra_headers.items()]
+            if "metadata" in request_options:
+                request_options["metadata"].extend(metadata)
+            else:
+                request_options["metadata"] = metadata
+
         try:
             if stream:
                 with generation_types.rewrite_stream_error():
@@ -351,6 +360,7 @@ async def generate_content_async(
         tools: content_types.FunctionLibraryType | None = None,
         tool_config: content_types.ToolConfigType | None = None,
         request_options: helper_types.RequestOptionsType | None = None,
+        extra_headers: dict[str, str] | None = None,
     ) -> generation_types.AsyncGenerateContentResponse:
         """The async version of `GenerativeModel.generate_content`."""
         if not contents:
@@ -373,6 +383,15 @@ async def generate_content_async(
         if request_options is None:
             request_options = {}
 
+        # Convert extra_headers to metadata format if provided
+        metadata = []
+        if extra_headers:
+            metadata = [(k, v) for k, v in extra_headers.items()]
+
+        # Add metadata to request_options
+        if metadata:
+            request_options["metadata"] = metadata
+
         try:
             if stream:
                 with generation_types.rewrite_stream_error():
diff --git a/tests/test_generative_models.py b/tests/test_generative_models.py
index 74469e5b8..b9842e8d9 100644
--- a/tests/test_generative_models.py
+++ b/tests/test_generative_models.py
@@ -48,10 +48,16 @@ def __init__(self, test):
     def generate_content(
         self,
         request: protos.GenerateContentRequest,
+        *,
+        extra_headers: dict[str, str] | None = None,
         **kwargs,
     ) -> protos.GenerateContentResponse:
         self.test.assertIsInstance(request, protos.GenerateContentRequest)
         self.observed_requests.append(request)
+        # Convert extra_headers to metadata format and ensure it's in kwargs
+        if extra_headers:
+            metadata = [(k, v) for k, v in extra_headers.items()]
+            kwargs.setdefault("metadata", []).extend(metadata) 
         self.observed_kwargs.append(kwargs)
         response = self.responses["generate_content"].pop(0)
         return response
@@ -59,9 +65,15 @@ def generate_content(
     def stream_generate_content(
         self,
         request: protos.GetModelRequest,
+        *,
+        extra_headers: dict[str, str] | None = None,
         **kwargs,
     ) -> Iterable[protos.GenerateContentResponse]:
         self.observed_requests.append(request)
+        # Convert extra_headers to metadata format and ensure it's in kwargs
+        if extra_headers:
+            metadata = [(k, v) for k, v in extra_headers.items()]
+            kwargs.setdefault("metadata", []).extend(metadata)
         self.observed_kwargs.append(kwargs)
         response = self.responses["stream_generate_content"].pop(0)
         return response
diff --git a/tests/test_generative_models_async.py b/tests/test_generative_models_async.py
index b37c65235..5a26d4914 100644
--- a/tests/test_generative_models_async.py
+++ b/tests/test_generative_models_async.py
@@ -53,19 +53,35 @@ def add_client_method(f):
         @add_client_method
         async def generate_content(
             request: protos.GenerateContentRequest,
+            *,
+            extra_headers: dict[str, str] | None = None,
             **kwargs,
         ) -> protos.GenerateContentResponse:
             self.assertIsInstance(request, protos.GenerateContentRequest)
             self.observed_requests.append(request)
+
+            if extra_headers:
+                metadata = [(k, v) for k, v in extra_headers.items()]
+                kwargs.setdefault("metadata", []).extend(metadata)  # Merge with existing metadata if any
+
+            self.observed_kwargs.append(kwargs)
             response = self.responses["generate_content"].pop(0)
             return response
 
         @add_client_method
         async def stream_generate_content(
             request: protos.GetModelRequest,
+            *,
+            extra_headers: dict[str, str] | None = None,
             **kwargs,
         ) -> Iterable[protos.GenerateContentResponse]:
             self.observed_requests.append(request)
+            # Convert extra_headers to metadata format and ensure it's in kwargs
+            if extra_headers:
+                metadata = [(k, v) for k, v in extra_headers.items()]
+                kwargs.setdefault("metadata", []).extend(metadata)
+
+            self.observed_kwargs.append(kwargs)
             response = self.responses["stream_generate_content"].pop(0)
             return response