diff --git a/litellm/llms/openai/openai.py b/litellm/llms/openai/openai.py
index 87c944a642a..f330d7fc6d4 100644
--- a/litellm/llms/openai/openai.py
+++ b/litellm/llms/openai/openai.py
@@ -1,3 +1,4 @@
+import inspect
 import time
 import types
 from collections.abc import AsyncIterator, Callable, Coroutine, Iterable, Iterator, Mapping, Sequence
@@ -379,6 +380,11 @@ _EXTRA_HEADERS_ADAPTER: Final = TypeAdapter(dict[str, str] | None)
 _EXTRA_QUERY_ADAPTER: Final = TypeAdapter(dict[str, object] | None)
 _NO_EXTRA_HEADERS: Final[Mapping[str, str]] = types.MappingProxyType({})
 _SDK_OPTION_KEYS: Final = frozenset(("extra_headers", "extra_query", "extra_body"))
+_CLIENT_API_KEY_ONLY_AUTH: Final[Mapping[str, Mapping[str, bool]]] = (
+    types.MappingProxyType({"security": types.MappingProxyType({"bearer_auth": True})})
+    if "security" in inspect.signature(make_request_options).parameters
+    else types.MappingProxyType({})
+)
 
 
 def _raw_chat_response(
@@ -398,6 +404,7 @@ def _request_without_sdk_transform(
         extra_query=_EXTRA_QUERY_ADAPTER.validate_python(data.get("extra_query")),
         extra_body=data.get("extra_body"),
         timeout=timeout,
+        **_CLIENT_API_KEY_ONLY_AUTH,  # pyright: ignore[reportArgumentType]  # openai<3 make_request_options has no security kwarg
     )
     return body, options
 
diff --git a/tests/unit/llms/openai/test_openai.py b/tests/unit/llms/openai/test_openai.py
index f61cfba9a10..53122246de4 100644
--- a/tests/unit/llms/openai/test_openai.py
+++ b/tests/unit/llms/openai/test_openai.py
@@ -1,6 +1,8 @@
 import asyncio
+import contextlib
 import gzip
 import importlib
+import inspect
 import json
 import os
 from collections.abc import Awaitable, Callable, Mapping, Sequence
@@ -227,6 +229,52 @@ async def test_acompletion_returns_prompt_cache_details_over_injected_transport(
     assert prompt_tokens_details.cached_tokens == 7
 
 
+_SDK_READS_OPENAI_ADMIN_KEY: Final = "admin_api_key" in inspect.signature(AsyncOpenAI.__init__).parameters
+
+
+@pytest.mark.skipif(not _SDK_READS_OPENAI_ADMIN_KEY, reason="openai<3 has no OPENAI_ADMIN_KEY fallback to leak")
+@pytest.mark.asyncio
+async def test_acompletion_never_sends_the_openai_admin_key_for_a_deployment_without_a_key(
+    monkeypatch: pytest.MonkeyPatch,
+):
+    monkeypatch.delenv("OPENAI_API_KEY", raising=False)
+    monkeypatch.setenv("OPENAI_ADMIN_KEY", "sk-admin-only")
+    monkeypatch.setattr(litellm, "api_key", None)
+    monkeypatch.setattr(litellm, "openai_key", None)
+    outbound: Final = asyncio.Queue()
+
+    def respond(request: httpx.Request) -> httpx.Response:
+        outbound.put_nowait(request.headers.get("authorization"))
+        return httpx.Response(
+            200,
+            json={
+                "id": "chatcmpl-no-admin-key",
+                "object": "chat.completion",
+                "created": 0,
+                "model": "gpt-5.4-mini",
+                "choices": [{"index": 0, "message": {"role": "assistant", "content": "ok"}, "finish_reason": "stop"}],
+                "usage": {"prompt_tokens": 1, "completion_tokens": 1, "total_tokens": 2},
+            },
+        )
+
+    async with httpx.AsyncClient(transport=httpx.MockTransport(respond)) as http_client:
+        monkeypatch.setattr(litellm, "aclient_session", http_client)
+        with contextlib.suppress(litellm.AuthenticationError, litellm.InternalServerError):
+            await asyncio.wait_for(
+                litellm.acompletion(
+                    model="openai/gpt-5.4-mini",
+                    api_key="",
+                    messages=[{"role": "user", "content": "who signs this request"}],
+                    num_retries=0,
+                    max_retries=0,
+                ),
+                timeout=10,
+            )
+
+    sent: Final = tuple(outbound.get_nowait() for _ in range(outbound.qsize()))
+    assert all(authorization is None for authorization in sent), sent
+
+
 
 class _SearchingChatGate(CustomLogger):
     def __init__(self) -> None: