diff --git a/litellm/tests/test_exceptions.py b/litellm/tests/test_exceptions.py
index 13ad42b503c..2e0d79f16de 100644
--- a/litellm/tests/test_exceptions.py
+++ b/litellm/tests/test_exceptions.py
@@ -536,6 +536,55 @@ def test_completion_openai_api_key_exception():
 
 # tesy_async_acompletion()
 
+
+def test_router_completion_vertex_exception():
+    try:
+        import litellm
+
+        litellm.set_verbose = True
+        router = litellm.Router(
+            model_list=[
+                {
+                    "model_name": "vertex-gemini-pro",
+                    "litellm_params": {
+                        "model": "vertex_ai/gemini-pro",
+                        "api_key": "good-morning",
+                    },
+                },
+            ]
+        )
+        response = router.completion(
+            model="vertex-gemini-pro",
+            messages=[{"role": "user", "content": "hello"}],
+            vertex_project="bad-project",
+        )
+        pytest.fail("Request should have failed - bad api key")
+    except Exception as e:
+        print("exception: ", e)
+        assert "model: vertex_ai/gemini-pro" in str(e)
+        assert "model_group: vertex-gemini-pro" in str(e)
+        assert "deployment: vertex_ai/gemini-pro" in str(e)
+
+
+def test_litellm_completion_vertex_exception():
+    try:
+        import litellm
+
+        litellm.set_verbose = True
+        response = completion(
+            model="vertex_ai/gemini-pro",
+            api_key="good-morning",
+            messages=[{"role": "user", "content": "hello"}],
+            vertex_project="bad-project",
+        )
+        pytest.fail("Request should have failed - bad api key")
+    except Exception as e:
+        print("exception: ", e)
+        assert "model: vertex_ai/gemini-pro" in str(e)
+        assert "model_group" not in str(e)
+        assert "deployment" not in str(e)
+
+
 # # test_invalid_request_error(model="command-nightly")
 # # Test 3: Rate Limit Errors
 # def test_model_call(model):
diff --git a/litellm/utils.py b/litellm/utils.py
index e10a44e1aa7..47fbffa4702 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -7812,6 +7812,19 @@ def exception_type(
                         response=original_exception.response,
                     )
             elif custom_llm_provider == "vertex_ai":
+                if completion_kwargs is not None:
+                    # add model, deployment and model_group to the exception message
+                    _model = completion_kwargs.get("model")
+                    _kwargs = completion_kwargs.get("kwargs", {}) or {}
+                    _metadata = _kwargs.get("metadata", {}) or {}
+                    _model_group = _metadata.get("model_group")
+                    _deployment = _metadata.get("deployment")
+                    error_str += f"\nmodel: {_model}\n"
+                    if _model_group is not None:
+                        error_str += f"model_group: {_model_group}\n"
+                    if _deployment is not None:
+                        error_str += f"deployment: {_deployment}\n"
+
                 if (
                     "Vertex AI API has not been used in project" in error_str
                     or "Unable to find your project" in error_str