diff --git a/litellm/proxy/image_endpoints/endpoints.py b/litellm/proxy/image_endpoints/endpoints.py
index 28fb9baf2aa..84670f8c6dd 100644
--- a/litellm/proxy/image_endpoints/endpoints.py
+++ b/litellm/proxy/image_endpoints/endpoints.py
@@ -1,4 +1,5 @@
 import asyncio
+from datetime import datetime
 import io
 from collections.abc import Sequence
 from itertools import chain
@@ -132,6 +133,16 @@ async def image_generation(
         if data["model"] in litellm.model_alias_map:
             data["model"] = litellm.model_alias_map[data["model"]]
 
+        ### LOGGING OBJECT SETUP ### - ensures router rejected requests are dispatched to failure callbacks
+        start_time: Final = datetime.now()
+        logging_obj, data = litellm.utils.function_setup(
+            original_function="aimage_generation",
+            rules_obj=litellm.utils.Rules(),
+            start_time=start_time,
+            **data,
+        )
+        data["litellm_logging_obj"] = logging_obj
+
         ### CALL HOOKS ### - modify incoming data / reject request before calling the model
         prompt_value: Final = data.get("prompt")
         if prompt_value is not None:
diff --git a/tests/unit/proxy/image_endpoints/test_endpoints.py b/tests/unit/proxy/image_endpoints/test_endpoints.py
index f4aebecc11e..51502cbec03 100644
--- a/tests/unit/proxy/image_endpoints/test_endpoints.py
+++ b/tests/unit/proxy/image_endpoints/test_endpoints.py
@@ -116,6 +116,7 @@ async def test_image_generation_prompt_rerouting(monkeypatch):
     assert pre_call_input["messages"][0]["content"] == "original prompt"
     assert captured_route_request_data["prompt"] == "sanitized prompt"
     assert "messages" not in captured_route_request_data
+    assert captured_route_request_data.get("litellm_logging_obj") is not None
     assert response.headers.get("x-callback-test") == "value"