diff --git a/litellm/utils.py b/litellm/utils.py
index 796547cb042..6be1af4271a 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -7941,6 +7941,16 @@ def validate_and_fix_openai_messages(messages: List):
         if message.get("tool_calls"):
             message["tool_calls"] = jsonify_tools(tools=message["tool_calls"])
 
+        content = message.get("content")
+        if isinstance(content, list):
+            normalized_content = []
+            for item in content:
+                if isinstance(item, str):
+                    normalized_content.append({"type": "text", "text": item})
+                else:
+                    normalized_content.append(item)
+            message["content"] = normalized_content
+
         convert_msg_to_dict = cast(AllMessageValues, convert_to_dict(message))
         cleaned_message = cleanup_none_field_in_message(message=convert_msg_to_dict)
         new_messages.append(cleaned_message)
diff --git a/tests/test_litellm/test_utils.py b/tests/test_litellm/test_utils.py
index bc60375f906..209831783b8 100644
--- a/tests/test_litellm/test_utils.py
+++ b/tests/test_litellm/test_utils.py
@@ -27,6 +27,7 @@ from litellm.utils import (
     get_llm_provider,
     get_optional_params_image_gen,
     is_cached_message,
+    validate_and_fix_openai_messages,
 )
 
 # Adds the parent directory to the system path
@@ -3277,6 +3278,37 @@ class TestIsCachedMessage:
         assert is_cached_message(message) is False
 
 
+def test_normalize_array_of_strings_in_content():
+    """String items in list content become OpenAI multimodal text parts; dict parts unchanged."""
+    only_strings = validate_and_fix_openai_messages(
+        [
+            {
+                "role": "user",
+                "content": ["what is the capital of France?"],
+            }
+        ]
+    )
+    assert only_strings[0]["content"] == [
+        {"type": "text", "text": "what is the capital of France?"}
+    ]
+
+    mixed = validate_and_fix_openai_messages(
+        [
+            {
+                "role": "user",
+                "content": [
+                    "some text",
+                    {"type": "image_url", "image_url": {"url": "https://example.com/x.png"}},
+                ],
+            }
+        ]
+    )
+    assert mixed[0]["content"] == [
+        {"type": "text", "text": "some text"},
+        {"type": "image_url", "image_url": {"url": "https://example.com/x.png"}},
+    ]
+
+
 @pytest.mark.asyncio
 class TestProxyLoggingBudgetAlerts:
     """Test budget_alerts method in ProxyLogging class."""
@@ -4107,3 +4139,18 @@ class TestValidateAndFixThinkingParam:
         validate_and_fix_thinking_param(thinking=thinking)
         assert "budgetTokens" in thinking
         assert "budget_tokens" not in thinking
+
+
+def test_normalize_array_of_strings_in_content():
+    from litellm.utils import validate_and_fix_openai_messages
+
+    messages = [
+        {
+            "role": "user",
+            "content": ["what is the capital of France?"],
+        }
+    ]
+    result = validate_and_fix_openai_messages(messages=messages)
+    assert result[0]["content"] == [
+        {"type": "text", "text": "what is the capital of France?"},
+    ]
\ No newline at end of file