diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 49d3fbc88c..f2d59b15fb 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -2800,7 +2800,9 @@ async def chat_completion(
 
         ## LOGGING OBJECT ## - initialize logging object for logging success/failure events for call
         ## IMPORTANT Note: - initialize this before running pre-call checks. Ensures we log rejected requests to langfuse.
-        data["litellm_call_id"] = request.headers.get('x-litellm-call-id', str(uuid.uuid4()))
+        data["litellm_call_id"] = request.headers.get(
+            "x-litellm-call-id", str(uuid.uuid4())
+        )
         logging_obj, data = litellm.utils.function_setup(
             original_function="acompletion",
             rules_obj=litellm.utils.Rules(),
@@ -3137,6 +3139,7 @@ async def completion(
         cache_key = hidden_params.get("cache_key", None) or ""
         api_base = hidden_params.get("api_base", None) or ""
         response_cost = hidden_params.get("response_cost", None) or ""
+        litellm_call_id = hidden_params.get("litellm_call_id", None) or ""
 
         ### ALERTING ###
         asyncio.create_task(
@@ -3151,7 +3154,7 @@ async def completion(
         ):  # use generate_responses to stream responses
             custom_headers = get_custom_headers(
                 user_api_key_dict=user_api_key_dict,
-                call_id=logging_obj.litellm_call_id,
+                call_id=litellm_call_id,
                 model_id=model_id,
                 cache_key=cache_key,
                 api_base=api_base,
@@ -3172,7 +3175,7 @@ async def completion(
         fastapi_response.headers.update(
             get_custom_headers(
                 user_api_key_dict=user_api_key_dict,
-                call_id=logging_obj.litellm_call_id,
+                call_id=litellm_call_id,
                 model_id=model_id,
                 cache_key=cache_key,
                 api_base=api_base,
@@ -3391,6 +3394,7 @@ async def embeddings(
         cache_key = hidden_params.get("cache_key", None) or ""
         api_base = hidden_params.get("api_base", None) or ""
         response_cost = hidden_params.get("response_cost", None) or ""
+        litellm_call_id = hidden_params.get("litellm_call_id", None) or ""
 
         fastapi_response.headers.update(
             get_custom_headers(
@@ -3401,6 +3405,7 @@ async def embeddings(
                 version=version,
                 response_cost=response_cost,
                 model_region=getattr(user_api_key_dict, "allowed_model_region", ""),
+                call_id=litellm_call_id,
             )
         )
 
@@ -3545,6 +3550,7 @@ async def image_generation(
         cache_key = hidden_params.get("cache_key", None) or ""
         api_base = hidden_params.get("api_base", None) or ""
         response_cost = hidden_params.get("response_cost", None) or ""
+        litellm_call_id = hidden_params.get("litellm_call_id", None) or ""
 
         fastapi_response.headers.update(
             get_custom_headers(
@@ -3555,6 +3561,7 @@ async def image_generation(
                 version=version,
                 response_cost=response_cost,
                 model_region=getattr(user_api_key_dict, "allowed_model_region", ""),
+                call_id=litellm_call_id,
             )
         )
 
@@ -3686,6 +3693,7 @@ async def audio_speech(
         cache_key = hidden_params.get("cache_key", None) or ""
         api_base = hidden_params.get("api_base", None) or ""
         response_cost = hidden_params.get("response_cost", None) or ""
+        litellm_call_id = hidden_params.get("litellm_call_id", None) or ""
 
         # Printing each chunk size
         async def generate(_response: HttpxBinaryResponseContent):
@@ -3702,6 +3710,7 @@ async def audio_speech(
             response_cost=response_cost,
             model_region=getattr(user_api_key_dict, "allowed_model_region", ""),
             fastest_response_batch_completion=None,
+            call_id=litellm_call_id,
         )
 
         selected_data_generator = select_data_generator(
@@ -3856,6 +3865,7 @@ async def audio_transcriptions(
         cache_key = hidden_params.get("cache_key", None) or ""
         api_base = hidden_params.get("api_base", None) or ""
         response_cost = hidden_params.get("response_cost", None) or ""
+        litellm_call_id = hidden_params.get("litellm_call_id", None) or ""
 
         fastapi_response.headers.update(
             get_custom_headers(
@@ -3866,6 +3876,7 @@ async def audio_transcriptions(
                 version=version,
                 response_cost=response_cost,
                 model_region=getattr(user_api_key_dict, "allowed_model_region", ""),
+                call_id=litellm_call_id,
             )
         )
 
diff --git a/litellm/utils.py b/litellm/utils.py
index 65b4d63ebe..03ed52b5c0 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -1332,6 +1332,9 @@ def client(original_function):
 
             # ADD HIDDEN PARAMS - additional call metadata
             if hasattr(result, "_hidden_params"):
+                result._hidden_params["litellm_call_id"] = getattr(
+                    logging_obj, "litellm_call_id", None
+                )
                 result._hidden_params["model_id"] = kwargs.get("model_info", {}).get(
                     "id", None
                 )