diff --git a/libs/partners/openai/langchain_openai/chat_models/base.py b/libs/partners/openai/langchain_openai/chat_models/base.py
index 27d4adf06817a..4e91fab34a8d5 100644
--- a/libs/partners/openai/langchain_openai/chat_models/base.py
+++ b/libs/partners/openai/langchain_openai/chat_models/base.py
@@ -2161,16 +2161,18 @@ def _create_usage_metadata(oai_token_usage: dict) -> UsageMetadata:
     output_tokens = oai_token_usage.get("completion_tokens", 0)
     total_tokens = oai_token_usage.get("total_tokens", input_tokens + output_tokens)
     input_token_details: dict = {
-        "audio": oai_token_usage.get("prompt_tokens_details", {}).get("audio_tokens"),
-        "cache_read": oai_token_usage.get("prompt_tokens_details", {}).get(
+        "audio": (oai_token_usage.get("prompt_tokens_details") or {}).get(
+            "audio_tokens"
+        ),
+        "cache_read": (oai_token_usage.get("prompt_tokens_details") or {}).get(
             "cached_tokens"
         ),
     }
     output_token_details: dict = {
-        "audio": oai_token_usage.get("completion_tokens_details", {}).get(
+        "audio": (oai_token_usage.get("completion_tokens_details") or {}).get(
             "audio_tokens"
         ),
-        "reasoning": oai_token_usage.get("completion_tokens_details", {}).get(
+        "reasoning": (oai_token_usage.get("completion_tokens_details") or {}).get(
             "reasoning_tokens"
         ),
     }
diff --git a/libs/partners/openai/tests/unit_tests/chat_models/test_base.py b/libs/partners/openai/tests/unit_tests/chat_models/test_base.py
index d205276704016..cc03698e5ef93 100644
--- a/libs/partners/openai/tests/unit_tests/chat_models/test_base.py
+++ b/libs/partners/openai/tests/unit_tests/chat_models/test_base.py
@@ -23,6 +23,7 @@
 from langchain_openai.chat_models.base import (
     _convert_dict_to_message,
     _convert_message_to_dict,
+    _create_usage_metadata,
     _format_message_content,
 )
 
@@ -730,3 +731,21 @@ def test_schema_from_with_structured_output(schema: Type) -> None:
     }
     actual = structured_llm.get_output_schema().model_json_schema()
     assert actual == expected
+
+
+def test__create_usage_metadata() -> None:
+    usage_metadata = {
+        "completion_tokens": 15,
+        "prompt_tokens_details": None,
+        "completion_tokens_details": None,
+        "prompt_tokens": 11,
+        "total_tokens": 26,
+    }
+    result = _create_usage_metadata(usage_metadata)
+    assert result == UsageMetadata(
+        output_tokens=15,
+        input_tokens=11,
+        total_tokens=26,
+        input_token_details={},
+        output_token_details={},
+    )