diff --git a/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py b/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py
index 255c4a0902e..6992f9fb015 100644
--- a/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py
+++ b/litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py
@@ -2503,6 +2503,8 @@ class VertexGeminiConfig(VertexAIBaseConfig, BaseConfig):
             VertexGeminiConfig._set_grounding_usage_counters(usage, grounding_metadata)
 
             setattr(model_response, "usage", usage)
+            if isinstance(cached_content_creation, dict):
+                model_response._hidden_params[VERTEX_AI_CACHED_CONTENT_KEY] = cached_content_creation
 
             ## ADD METADATA TO RESPONSE ##
 
@@ -3265,6 +3267,9 @@ class ModelResponseIterator:
 
         VertexGeminiConfig._set_grounding_usage_counters(usage, grounding_metadata)
 
+        if isinstance(cached_content_creation, dict):
+            model_response._hidden_params[VERTEX_AI_CACHED_CONTENT_KEY] = cached_content_creation
+
         traffic_type: Final = processed_chunk.get("usageMetadata", {}).get("trafficType")
         if traffic_type:
             model_response._hidden_params.setdefault("provider_specific_fields", {})["traffic_type"] = traffic_type
diff --git a/tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py b/tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py
index d31322c9ba4..35c827fff2a 100644
--- a/tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py
+++ b/tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py
@@ -6429,6 +6429,14 @@ def test_transform_response_applies_cache_creation_usage(include_creation):
     if include_creation:
         assert result.usage.prompt_tokens == 20010
         assert result.usage.cache_creation_input_tokens == 10000
+        assert result._hidden_params["vertex_ai_cached_content"] == {
+            "name": "cached-content",
+            "model": model,
+            "total_token_count": 10000,
+            "create_time": None,
+            "expire_time": None,
+        }
     else:
         assert result.usage.prompt_tokens == 10010
         assert not hasattr(result.usage, "cache_creation_input_tokens")
+        assert "vertex_ai_cached_content" not in result._hidden_params