Enhance error handling in FileContentStreamingHandler for custom LLM provider routing

- Added validation to ensure credentials include a custom LLM provider before routing.
- Cleaned up type casting for better readability.
- Introduced a new test to verify behavior when a non-OpenAI provider is used, ensuring proper handling of streaming responses.
- Updated imports to include necessary modules for testing.

litellm/proxy/openai_files_endpoints/file_content_streaming_handler.py

@@ -64,14 +64,16 @@ class FileContentStreamingHandler:
     ) -> StreamingResponse:
         effective_custom_llm_provider = custom_llm_provider
         if should_route:
+            if credentials is None or credentials.get("custom_llm_provider") is None:
+                raise ValueError(
+                    "Model-based file routing requires credentials with custom_llm_provider"
+                )
             prepare_data_with_credentials(
                 data=data,
-                credentials=credentials,  # type: ignore[arg-type]
+                credentials=credentials,
                 file_id=original_file_id,
             )
-            effective_custom_llm_provider = cast(
-                str, credentials["custom_llm_provider"]
-            )
+            effective_custom_llm_provider = cast(str, credentials["custom_llm_provider"])
 
         stream_result = cast(
             FileContentStreamingResult,

tests/test_litellm/proxy/openai_files_endpoint/test_files_endpoint.py

@@ -5,6 +5,7 @@ from unittest.mock import ANY, AsyncMock
 
 import pytest
 import respx
+import httpx
 from fastapi.testclient import TestClient
 from pytest_mock import MockerFixture
 
@@ -22,7 +23,7 @@ from litellm.proxy.openai_files_endpoints.file_content_streaming_handler import
     FileContentStreamingHandler,
 )
 from litellm.proxy.proxy_server import app
-from litellm.types.llms.openai import OpenAIFileObject
+from litellm.types.llms.openai import HttpxBinaryResponseContent, OpenAIFileObject
 
 client = TestClient(app)
 from litellm.caching.caching import DualCache
@@ -1713,3 +1714,67 @@ def test_get_file_content_streams_with_routed_provider(
     assert captured_kwargs["stream"] is True
     proxy_logging_obj.update_request_status.assert_awaited_once()
     proxy_logging_obj.post_call_failure_hook.assert_not_called()
+
+
+def test_get_file_content_non_openai_provider_skips_streaming_handler(
+    mocker: MockerFixture, monkeypatch, llm_router: Router
+):
+    import litellm.proxy.proxy_server as ps
+    from litellm.proxy._types import LitellmUserRoles
+
+    proxy_logging_obj = setup_proxy_logging_object(monkeypatch, llm_router)
+    monkeypatch.setattr("litellm.proxy.proxy_server.master_key", None)
+    monkeypatch.setattr("litellm.proxy.proxy_server.prisma_client", None)
+    proxy_logging_obj.post_call_failure_hook = mocker.AsyncMock()
+
+    captured_kwargs = {}
+
+    async def _mock_afile_content(**kwargs):
+        captured_kwargs.update(kwargs)
+        return HttpxBinaryResponseContent(
+            response=httpx.Response(
+                status_code=200,
+                content=b"azure-bytes",
+                headers={
+                    "content-type": "application/octet-stream",
+                    "content-length": "11",
+                },
+            )
+        )
+
+    mock_streaming_response = mocker.AsyncMock()
+
+    monkeypatch.setattr(litellm, "afile_content", _mock_afile_content)
+    monkeypatch.setattr(
+        FileContentStreamingHandler,
+        "get_streaming_file_content_response",
+        mock_streaming_response,
+    )
+    monkeypatch.setattr(
+        "litellm.proxy.openai_files_endpoints.files_endpoints.handle_model_based_routing",
+        lambda **kwargs: (False, None, None, None),
+    )
+
+    app.dependency_overrides[ps.user_api_key_auth] = lambda: UserAPIKeyAuth(
+        api_key="test-key",
+        user_role=LitellmUserRoles.PROXY_ADMIN,
+        user_id="test-user",
+    )
+
+    try:
+        response = client.get(
+            "/v1/files/file-abc123/content",
+            headers={
+                "Authorization": "Bearer test-key",
+                "custom-llm-provider": "azure",
+            },
+        )
+    finally:
+        app.dependency_overrides.pop(ps.user_api_key_auth, None)
+
+    assert response.status_code == 200, response.text
+    assert response.content == b"azure-bytes"
+    assert captured_kwargs["custom_llm_provider"] == "azure"
+    assert "stream" not in captured_kwargs
+    mock_streaming_response.assert_not_awaited()
+    proxy_logging_obj.post_call_failure_hook.assert_not_called()