diff --git a/litellm/llms/bedrock.py b/litellm/llms/bedrock.py
index 7014ebc42b5..8c9a98352a3 100644
--- a/litellm/llms/bedrock.py
+++ b/litellm/llms/bedrock.py
@@ -392,11 +392,9 @@ def completion(
 
     model_response["created"] = time.time()
     model_response["model"] = model
-    model_response["usage"] = {
-        "prompt_tokens": prompt_tokens,
-        "completion_tokens": completion_tokens,
-        "total_tokens": prompt_tokens + completion_tokens,
-    }
+    model_response.usage.completion_tokens = completion_tokens
+    model_response.usage.prompt_tokens = prompt_tokens
+    model_response.usage.total_tokens = prompt_tokens + completion_tokens
     return model_response
 
 
diff --git a/litellm/llms/palm.py b/litellm/llms/palm.py
index b4160b63b5e..79a91364943 100644
--- a/litellm/llms/palm.py
+++ b/litellm/llms/palm.py
@@ -157,11 +157,9 @@ def completion(
 
     model_response["created"] = time.time()
     model_response["model"] = "palm/" + model
-    model_response["usage"] = {
-        "prompt_tokens": prompt_tokens,
-        "completion_tokens": completion_tokens,
-        "total_tokens": prompt_tokens + completion_tokens,
-    }
+    model_response.usage.completion_tokens = completion_tokens
+    model_response.usage.prompt_tokens = prompt_tokens
+    model_response.usage.total_tokens = prompt_tokens + completion_tokens
     return model_response
 
 def embedding():
diff --git a/litellm/llms/petals.py b/litellm/llms/petals.py
index a3127eade71..5834129c199 100644
--- a/litellm/llms/petals.py
+++ b/litellm/llms/petals.py
@@ -176,11 +176,9 @@ def completion(
 
     model_response["created"] = time.time()
     model_response["model"] = model
-    model_response["usage"] = {
-        "prompt_tokens": prompt_tokens,
-        "completion_tokens": completion_tokens,
-        "total_tokens": prompt_tokens + completion_tokens,
-    }
+    model_response.usage.completion_tokens = completion_tokens
+    model_response.usage.prompt_tokens = prompt_tokens
+    model_response.usage.total_tokens = prompt_tokens + completion_tokens
     return model_response
 
 def embedding():
diff --git a/litellm/llms/sagemaker.py b/litellm/llms/sagemaker.py
index 962a2fc0e1e..8c999af6396 100644
--- a/litellm/llms/sagemaker.py
+++ b/litellm/llms/sagemaker.py
@@ -169,11 +169,9 @@ def completion(
 
     model_response["created"] = time.time()
     model_response["model"] = model
-    model_response["usage"] = {
-        "prompt_tokens": prompt_tokens,
-        "completion_tokens": completion_tokens,
-        "total_tokens": prompt_tokens + completion_tokens,
-    }
+    model_response.usage.completion_tokens = completion_tokens
+    model_response.usage.prompt_tokens = prompt_tokens
+    model_response.usage.total_tokens = prompt_tokens + completion_tokens
     return model_response
 
 def embedding():