diff --git a/litellm/llms/azure/chat/gpt_5_transformation.py b/litellm/llms/azure/chat/gpt_5_transformation.py
index 2d0ce8b5bc..87f81d117f 100644
--- a/litellm/llms/azure/chat/gpt_5_transformation.py
+++ b/litellm/llms/azure/chat/gpt_5_transformation.py
@@ -60,7 +60,6 @@ class AzureOpenAIGPT5Config(AzureOpenAIConfig, OpenAIGPT5Config):
                     message=(
                         "Azure OpenAI does not support reasoning_effort='none' for this model. "
                         "Supported values are: 'low', 'medium', and 'high'. "
-                        "Note: gpt-5.1 does support reasoning_effort='none'. "
                         "To drop this parameter, set `litellm.drop_params=True` or for proxy:\n\n"
                         "`litellm_settings:\n drop_params: true`\n"
                         "Issue: https://github.com/BerriAI/litellm/issues/16704"
diff --git a/litellm/responses/utils.py b/litellm/responses/utils.py
index 198182cf11..ad99609e90 100644
--- a/litellm/responses/utils.py
+++ b/litellm/responses/utils.py
@@ -24,7 +24,12 @@ from litellm.types.llms.openai import (
     ResponseText,
 )
 from litellm.types.responses.main import DecodedResponseId
-from litellm.types.utils import PromptTokensDetails, SpecialEnums, Usage
+from litellm.types.utils import (
+    CompletionTokensDetailsWrapper,
+    PromptTokensDetails,
+    SpecialEnums,
+    Usage,
+)
 
 
 class ResponsesAPIRequestUtils:
@@ -366,10 +371,11 @@ class ResponsesAPIRequestUtils:
         Headers from tools.headers in request body should be passed to MCP server.
         """
         from starlette.datastructures import Headers
+
         from litellm.proxy._experimental.mcp_server.auth.user_api_key_auth_mcp import (
             MCPRequestHandler,
         )
-        
+
         # Extract headers from secret_fields which contains the original request headers
         raw_headers_from_request: Optional[Dict[str, str]] = None
         if secret_fields and isinstance(secret_fields, dict):
@@ -445,11 +451,20 @@ class ResponseAPILoggingUtils:
                 cached_tokens=response_api_usage.input_tokens_details.cached_tokens,
                 audio_tokens=response_api_usage.input_tokens_details.audio_tokens,
             )
+        completion_tokens_details: Optional[CompletionTokensDetailsWrapper] = None
+        if response_api_usage.output_tokens_details:
+            completion_tokens_details = CompletionTokensDetailsWrapper(
+                reasoning_tokens=getattr(
+                    response_api_usage.output_tokens_details, "reasoning_tokens", None
+                )
+            )
+            
         chat_usage = Usage(
             prompt_tokens=prompt_tokens,
             completion_tokens=completion_tokens,
             total_tokens=prompt_tokens + completion_tokens,
             prompt_tokens_details=prompt_tokens_details,
+            completion_tokens_details=completion_tokens_details,
         )
 
         # Preserve cost attribute if it exists on ResponseAPIUsage