diff --git a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_integration.py b/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_integration.py
deleted file mode 100644
index 8f9acafa49d..00000000000
--- a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_integration.py
+++ /dev/null
@@ -1,214 +0,0 @@
-"""
-Test Vertex AI files integration with main files API
-"""
-
-import pytest
-from unittest.mock import AsyncMock, MagicMock, patch
-
-import litellm
-from litellm.types.llms.openai import HttpxBinaryResponseContent
-
-
-class TestVertexAIFilesIntegration:
-    """Test integration of Vertex AI files with main litellm API"""
-
-    @pytest.mark.asyncio
-    async def test_litellm_afile_content_vertex_ai_provider(self):
-        """Test litellm.afile_content with vertex_ai provider"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        # Create a mock HttpxBinaryResponseContent response
-        import httpx
-
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-        )
-        mock_result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Mock the base_llm_http_handler.retrieve_file_content since the code
-        # now routes through ProviderConfigManager -> base_llm_http_handler
-        with patch(
-            "litellm.files.main.base_llm_http_handler.retrieve_file_content",
-            new_callable=MagicMock,
-        ) as mock_retrieve:
-            # Make it return a coroutine for async path
-            mock_retrieve.return_value = mock_result
-
-            result = await litellm.afile_content(
-                file_id=file_id,
-                custom_llm_provider="vertex_ai",
-                vertex_project="test-project",
-                vertex_location="us-central1",
-                vertex_credentials=None,
-            )
-
-            # Verify the result
-            assert isinstance(result, HttpxBinaryResponseContent)
-            assert result.response.content == expected_content
-            assert result.response.status_code == 200
-
-            # Verify the mock was called
-            mock_retrieve.assert_called_once()
-
-    def test_litellm_file_content_vertex_ai_provider(self):
-        """Test litellm.file_content with vertex_ai provider (sync)"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        # Create a mock HttpxBinaryResponseContent response
-        import httpx
-
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-        )
-        mock_result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Mock the base_llm_http_handler.retrieve_file_content
-        with patch(
-            "litellm.files.main.base_llm_http_handler.retrieve_file_content",
-            return_value=mock_result,
-        ) as mock_retrieve:
-            result = litellm.file_content(
-                file_id=file_id,
-                custom_llm_provider="vertex_ai",
-                vertex_project="test-project",
-                vertex_location="us-central1",
-                vertex_credentials=None,
-            )
-
-            # Verify the result
-            assert isinstance(result, HttpxBinaryResponseContent)
-            assert result.response.content == expected_content
-            assert result.response.status_code == 200
-
-            # Verify the mock was called
-            mock_retrieve.assert_called_once()
-
-    def test_litellm_file_content_vertex_ai_with_model_provider_detection(self):
-        """Test litellm.file_content with model parameter for provider detection"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        # Create a mock HttpxBinaryResponseContent response
-        import httpx
-
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-        )
-        mock_result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Mock the base_llm_http_handler.retrieve_file_content
-        with patch(
-            "litellm.files.main.base_llm_http_handler.retrieve_file_content",
-            return_value=mock_result,
-        ):
-            # Mock get_llm_provider to return vertex_ai
-            with patch("litellm.files.main.get_llm_provider") as mock_get_provider:
-                mock_get_provider.return_value = (
-                    "vertex_ai/gemini-pro",
-                    "vertex_ai",
-                    None,
-                    None,
-                )
-
-                # Call litellm.file_content with model to trigger provider detection
-                result = litellm.file_content(
-                    file_id=file_id,
-                    model="vertex_ai/gemini-pro",
-                    vertex_project="test-project",
-                    vertex_location="us-central1",
-                )
-
-                # Verify the result
-                assert isinstance(result, HttpxBinaryResponseContent)
-                assert result.response.content == expected_content
-
-                # Verify provider detection was called
-                mock_get_provider.assert_called_once()
-
-    def test_litellm_file_content_vertex_ai_error_cases(self):
-        """Test error handling in vertex_ai file_content"""
-        # Test missing file_id - the VertexAI provider config's
-        # transform_file_content_request should handle empty file_id.
-        # Since the code now goes through base_llm_http_handler, we mock
-        # ProviderConfigManager to return None so it falls through to the
-        # old vertex_ai code path that validates file_id.
-        with patch(
-            "litellm.files.main.ProviderConfigManager.get_provider_files_config",
-            return_value=None,
-        ):
-            with pytest.raises(ValueError, match="file_id is required"):
-                litellm.file_content(
-                    file_id="",  # Empty file_id should cause error
-                    custom_llm_provider="vertex_ai",
-                    vertex_project="test-project",
-                )
-
-    def test_vertex_ai_provider_in_supported_providers_list(self):
-        """Test that vertex_ai is included in supported providers for file_content"""
-        # This test ensures the type annotations and error messages include vertex_ai
-
-        # Test that calling with unsupported provider raises appropriate error
-        with pytest.raises(Exception, match="unsupported_provider' is not a valid LlmProviders") as exc_info:
-            litellm.file_content(
-                file_id="test-file-id",
-                custom_llm_provider="unsupported_provider",  # This should fail
-            )
-
-        # The error message should mention supported providers including vertex_ai
-        error_message = str(exc_info.value)
-        assert "vertex_ai" in error_message or "supported" in error_message.lower()
-
-    @pytest.mark.asyncio
-    async def test_vertex_ai_file_content_with_timeout_and_retries(self):
-        """Test vertex_ai file_content with timeout and retry configuration"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        # Create a mock HttpxBinaryResponseContent response
-        import httpx
-
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-        )
-        mock_result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Mock the base_llm_http_handler.retrieve_file_content
-        with patch(
-            "litellm.files.main.base_llm_http_handler.retrieve_file_content",
-            new_callable=MagicMock,
-        ) as mock_retrieve:
-            mock_retrieve.return_value = mock_result
-
-            # Call with custom timeout and max_retries
-            result = await litellm.afile_content(
-                file_id=file_id,
-                custom_llm_provider="vertex_ai",
-                vertex_project="test-project",
-                vertex_location="us-central1",
-                timeout=120,
-                max_retries=5,
-            )
-
-            # Verify the result
-            assert isinstance(result, HttpxBinaryResponseContent)
-            assert result.response.content == expected_content
-
-            # Verify the mock was called
-            mock_retrieve.assert_called_once()
-            # Verify the timeout was passed through
-            call_kwargs = mock_retrieve.call_args.kwargs
-            assert call_kwargs["timeout"] == 120
diff --git a/tests/unit/llms/tencent/__init__.py b/tests/unit/llms/tencent/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/unit/llms/tencent/messages/__init__.py b/tests/unit/llms/tencent/messages/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/tencent/messages/test_tencent_anthropic_messages_transformation.py b/tests/unit/llms/tencent/messages/test_tencent_anthropic_messages_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/tencent/messages/test_tencent_anthropic_messages_transformation.py
rename to tests/unit/llms/tencent/messages/test_tencent_anthropic_messages_transformation.py
diff --git a/tests/unit/llms/together_ai/__init__.py b/tests/unit/llms/together_ai/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/unit/llms/together_ai/chat/__init__.py b/tests/unit/llms/together_ai/chat/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/together_ai/chat/test_together_ai_chat_transformation.py b/tests/unit/llms/together_ai/chat/test_together_ai_chat_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/together_ai/chat/test_together_ai_chat_transformation.py
rename to tests/unit/llms/together_ai/chat/test_together_ai_chat_transformation.py
diff --git a/tests/unit/llms/valkey/__init__.py b/tests/unit/llms/valkey/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/unit/llms/valkey/vector_stores/__init__.py b/tests/unit/llms/valkey/vector_stores/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/valkey/vector_stores/test_valkey_transformation.py b/tests/unit/llms/valkey/vector_stores/test_valkey_transformation.py
similarity index 97%
rename from tests/test_litellm/llms/valkey/vector_stores/test_valkey_transformation.py
rename to tests/unit/llms/valkey/vector_stores/test_valkey_transformation.py
index aa114f128c5..64e15008536 100644
--- a/tests/test_litellm/llms/valkey/vector_stores/test_valkey_transformation.py
+++ b/tests/unit/llms/valkey/vector_stores/test_valkey_transformation.py
@@ -1,8 +1,7 @@
 import struct
-import sys
 from types import SimpleNamespace
 from typing import Final
-from unittest.mock import MagicMock, patch
+from unittest.mock import MagicMock
 from urllib.parse import unquote, urlsplit
 
 import httpx
@@ -355,13 +354,6 @@ def test_search_treats_an_explicit_null_max_num_results_as_the_default():
     assert client.index.searched_query.query_string() == "*=>[KNN 10 @embedding $vec AS vector_distance]"
 
 
-def test_missing_redis_dependency_raises_actionable_error():
-    config = ValkeyVectorStoreConfig(sync_client=FakeRedis(), embedding_fn=FakeEmbeddingFn([1.0]))
-    blocked = {name: None for name in list(sys.modules) if name == "redis" or name.startswith("redis.")}
-
-    with patch.dict(sys.modules, blocked):
-        with pytest.raises(ValueError, match="pip install redis"):
-            _search(config)
 
 
 @pytest.mark.asyncio
diff --git a/tests/unit/llms/vercel_ai_gateway/__init__.py b/tests/unit/llms/vercel_ai_gateway/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/unit/llms/vercel_ai_gateway/chat/__init__.py b/tests/unit/llms/vercel_ai_gateway/chat/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vercel_ai_gateway/chat/test_vercel_ai_gateway_transformation.py b/tests/unit/llms/vercel_ai_gateway/chat/test_vercel_ai_gateway_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vercel_ai_gateway/chat/test_vercel_ai_gateway_transformation.py
rename to tests/unit/llms/vercel_ai_gateway/chat/test_vercel_ai_gateway_transformation.py
diff --git a/tests/unit/llms/vercel_ai_gateway/embedding/__init__.py b/tests/unit/llms/vercel_ai_gateway/embedding/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vercel_ai_gateway/embedding/test_vercel_ai_gateway_embedding.py b/tests/unit/llms/vercel_ai_gateway/embedding/test_vercel_ai_gateway_embedding.py
similarity index 100%
rename from tests/test_litellm/llms/vercel_ai_gateway/embedding/test_vercel_ai_gateway_embedding.py
rename to tests/unit/llms/vercel_ai_gateway/embedding/test_vercel_ai_gateway_embedding.py
diff --git a/tests/unit/llms/vertex_ai/__init__.py b/tests/unit/llms/vertex_ai/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/unit/llms/vertex_ai/agent_engine/__init__.py b/tests/unit/llms/vertex_ai/agent_engine/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/agent_engine/test_transformation.py b/tests/unit/llms/vertex_ai/agent_engine/test_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/agent_engine/test_transformation.py
rename to tests/unit/llms/vertex_ai/agent_engine/test_transformation.py
diff --git a/tests/unit/llms/vertex_ai/context_caching/__init__.py b/tests/unit/llms/vertex_ai/context_caching/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/context_caching/test_context_caching_ttl.py b/tests/unit/llms/vertex_ai/context_caching/test_context_caching_ttl.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/context_caching/test_context_caching_ttl.py
rename to tests/unit/llms/vertex_ai/context_caching/test_context_caching_ttl.py
diff --git a/tests/test_litellm/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py b/tests/unit/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py
similarity index 95%
rename from tests/test_litellm/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py
rename to tests/unit/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py
index 34c00e84d2e..4ba2d69a38b 100644
--- a/tests/test_litellm/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py
+++ b/tests/unit/llms/vertex_ai/context_caching/test_vertex_ai_context_caching.py
@@ -9,7 +9,6 @@ from litellm.litellm_core_utils.litellm_logging import Logging
 from litellm.llms.custom_httpx.http_handler import AsyncHTTPHandler, HTTPHandler
 from litellm.llms.vertex_ai.common_utils import VertexAIError
 from litellm.llms.vertex_ai.context_caching.vertex_ai_context_caching import (
-    MAX_PAGINATION_PAGES,
     ContextCachingEndpoints,
 )
 
@@ -1892,100 +1891,7 @@ class TestCheckCachePagination:
         assert result is None
         assert self.mock_async_client.get.call_count == 1
 
-    @pytest.mark.parametrize(
-        "custom_llm_provider", ["gemini", "vertex_ai", "vertex_ai_beta"]
-    )
-    @patch.object(ContextCachingEndpoints, "_get_token_and_url_context_caching")
-    def test_check_cache_pagination_max_pages_limit(
-        self, mock_get_token_url, custom_llm_provider
-    ):
-        """Test that pagination stops after MAX_PAGINATION_PAGES iterations"""
-        # Setup
-        mock_get_token_url.return_value = ("token", "https://test-url.com")
-        cache_key_to_find = "nonexistent_cache_key"
 
-        # Create mock response that always has nextPageToken (infinite pagination scenario)
-        def create_page_response(page_num):
-            response = MagicMock()
-            response.json.return_value = {
-                "cachedContents": [
-                    {"name": f"cache_{page_num}", "displayName": f"key_{page_num}"}
-                ],
-                "nextPageToken": f"token_page_{page_num + 1}",
-            }
-            return response
-
-        # Create MAX_PAGINATION_PAGES responses, each with a nextPageToken
-        self.mock_client.get.side_effect = [
-            create_page_response(i) for i in range(MAX_PAGINATION_PAGES)
-        ]
-
-        # Execute
-        result = self.context_caching.check_cache(
-            cache_key=cache_key_to_find,
-            client=self.mock_client,
-            headers={"Authorization": "Bearer token"},
-            api_key="test_key",
-            api_base=None,
-            logging_obj=self.mock_logging,
-            custom_llm_provider=custom_llm_provider,
-            vertex_project="test_project",
-            vertex_location="us-central1",
-            vertex_auth_header="Bearer test-token",
-        )
-
-        # Assert - should return None after exhausting all pages without finding match
-        assert result is None
-        # Verify exactly MAX_PAGINATION_PAGES API calls were made (not more)
-        assert self.mock_client.get.call_count == MAX_PAGINATION_PAGES
-
-    @pytest.mark.asyncio
-    @pytest.mark.parametrize(
-        "custom_llm_provider", ["gemini", "vertex_ai", "vertex_ai_beta"]
-    )
-    @patch.object(ContextCachingEndpoints, "_get_token_and_url_context_caching")
-    async def test_async_check_cache_pagination_max_pages_limit(
-        self, mock_get_token_url, custom_llm_provider
-    ):
-        """Test that async pagination stops after MAX_PAGINATION_PAGES iterations"""
-        # Setup
-        mock_get_token_url.return_value = ("token", "https://test-url.com")
-        cache_key_to_find = "nonexistent_cache_key"
-
-        # Create mock response that always has nextPageToken (infinite pagination scenario)
-        def create_page_response(page_num):
-            response = MagicMock()
-            response.json.return_value = {
-                "cachedContents": [
-                    {"name": f"cache_{page_num}", "displayName": f"key_{page_num}"}
-                ],
-                "nextPageToken": f"token_page_{page_num + 1}",
-            }
-            return response
-
-        # Create MAX_PAGINATION_PAGES responses, each with a nextPageToken
-        self.mock_async_client.get = AsyncMock(
-            side_effect=[create_page_response(i) for i in range(MAX_PAGINATION_PAGES)]
-        )
-
-        # Execute
-        result = await self.context_caching.async_check_cache(
-            cache_key=cache_key_to_find,
-            client=self.mock_async_client,
-            headers={"Authorization": "Bearer token"},
-            api_key="test_key",
-            api_base=None,
-            logging_obj=self.mock_logging,
-            custom_llm_provider=custom_llm_provider,
-            vertex_project="test_project",
-            vertex_location="us-central1",
-            vertex_auth_header="Bearer test-token",
-        )
-
-        # Assert - should return None after exhausting all pages without finding match
-        assert result is None
-        # Verify exactly MAX_PAGINATION_PAGES async API calls were made (not more)
-        assert self.mock_async_client.get.call_count == MAX_PAGINATION_PAGES
 
 
 class TestVertexAIGlobalLocation:
diff --git a/tests/unit/llms/vertex_ai/files/__init__.py b/tests/unit/llms/vertex_ai/files/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/files/test_file_retrieve_provider_routing.py b/tests/unit/llms/vertex_ai/files/test_file_retrieve_provider_routing.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/files/test_file_retrieve_provider_routing.py
rename to tests/unit/llms/vertex_ai/files/test_file_retrieve_provider_routing.py
diff --git a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py b/tests/unit/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py
similarity index 71%
rename from tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py
rename to tests/unit/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py
index d2ee9d7d659..f4aee11c140 100644
--- a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py
+++ b/tests/unit/llms/vertex_ai/files/test_vertex_ai_binary_file_upload.py
@@ -11,8 +11,6 @@ import io
 import json
 import pytest
 
-import httpx
-
 from litellm.llms.custom_httpx.llm_http_handler import AsyncHTTPHandler
 from litellm.llms.vertex_ai.files.transformation import VertexAIFilesConfig
 from litellm.types.llms.openai import CreateFileRequest
@@ -96,39 +94,6 @@ class TestVertexAIBinaryFileUpload:
         assert isinstance(transformed_request, bytes)
         assert transformed_request == mock_png_content
 
-    @pytest.mark.asyncio
-    async def test_http_handler_accepts_bytes_without_decoding(self):
-        """
-        Test that httpx correctly accepts binary data without decoding.
-
-        This test verifies that bytes can be passed to httpx's post/put methods
-        without needing UTF-8 decoding, which is the core of our fix.
-        """
-        # Create mock binary data with non-UTF-8 bytes
-        mock_binary_data = b"\x00\x01\x02\x03\xff\xfe\xfd\xc4\xe5\xf2"
-
-        # Test that httpx accepts bytes in the data parameter
-        # We're testing the behavior, not making an actual request
-
-        # Verify that attempting to decode would fail (proving it's binary)
-        with pytest.raises(UnicodeDecodeError):
-            mock_binary_data.decode("utf-8")
-
-        # Verify that httpx Request accepts bytes
-        try:
-            request = httpx.Request(
-                method="POST",
-                url="https://example.com/upload",
-                data=mock_binary_data,
-                headers={"Content-Type": "application/octet-stream"},
-            )
-            # If we get here, httpx accepts bytes - which is what we need
-            assert request.content == mock_binary_data
-        except Exception as e:
-            pytest.fail(f"httpx should accept bytes in data parameter: {e}")
-
-        # Document the expected behavior
-        assert isinstance(mock_binary_data, bytes), "Binary file data should remain as bytes"
 
     @pytest.mark.asyncio
     async def test_jsonl_file_upload_returns_streaming_body(self):
@@ -224,36 +189,3 @@ class TestVertexAIBinaryFileUpload:
             litellm_params={},
         )
         assert isinstance(result3, bytes)
-
-    def test_bytes_type_preservation_documentation(self):
-        """
-        Documentation test: Verify that bytes are the correct type for binary uploads.
-
-        This test documents the expected behavior:
-        - Binary files (PDF, images, etc.) should remain as bytes
-        - Text files (JSONL) should be strings
-        - httpx accepts both bytes and strings in the 'data' parameter
-        - bytes should NEVER be decoded to UTF-8 for binary files
-        """
-        # This is a documentation test - it always passes
-        # but serves as a reference for the expected behavior
-
-        expected_behavior = {
-            "binary_files": {
-                "input_type": "bytes",
-                "output_type": "bytes",
-                "examples": ["PDF", "PNG", "JPEG", "binary data"],
-                "http_method": "POST or PUT",
-                "encoding": "none - preserve raw bytes",
-            },
-            "text_files": {
-                "input_type": "str or bytes",
-                "output_type": "bytes",
-                "examples": ["JSONL", "CSV", "TXT"],
-                "http_method": "POST",
-                "encoding": "UTF-8",
-            },
-        }
-
-        assert expected_behavior["binary_files"]["encoding"] == "none - preserve raw bytes"
-        assert expected_behavior["text_files"]["encoding"] == "UTF-8"
diff --git a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_handler.py b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_handler.py
similarity index 75%
rename from tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_handler.py
rename to tests/unit/llms/vertex_ai/files/test_vertex_ai_files_handler.py
index 0a44f0a9a74..e0f0b7e5c0b 100644
--- a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_handler.py
+++ b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_handler.py
@@ -2,14 +2,11 @@
 Test Vertex AI files handler functionality
 """
 
-import asyncio
 import re
 from types import MappingProxyType
 import pytest
 from unittest.mock import AsyncMock, patch
 
-import httpx
-
 from litellm.llms.vertex_ai.files.handler import VertexAIFilesHandler
 from litellm.types.llms.openai import FileContentRequest, HttpxBinaryResponseContent
 
@@ -312,104 +309,3 @@ class TestVertexAIFilesHandler:
         assert isinstance(result, HttpxBinaryResponseContent)
         dynamic_params = mock_download.call_args.kwargs["standard_callback_dynamic_params"]
         assert dynamic_params["gcs_bucket_name"] == "my-model-bucket"
-
-    def test_file_content_sync_success(self):
-        """Test successful sync file content retrieval"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        file_content_request = FileContentRequest(file_id=file_id, extra_headers=None, extra_body=None)
-
-        # Create expected response
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-        )
-        expected_result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Mock asyncio.run to return our expected result
-        with patch("asyncio.run") as mock_run:
-            mock_run.return_value = expected_result
-
-            result = self.handler.file_content(
-                _is_async=False,
-                file_content_request=file_content_request,
-                api_base="",
-                vertex_credentials=None,
-                vertex_project="test-project",
-                vertex_location="us-central1",
-                timeout=60.0,
-                max_retries=3,
-            )
-
-            # Verify the result
-            assert result == expected_result
-
-            # Verify asyncio.run was called (indicating sync execution)
-            mock_run.assert_called_once()
-
-    @pytest.mark.asyncio
-    async def test_file_content_async_mode(self):
-        """Test async file content retrieval when _is_async=True"""
-        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
-        expected_content = b"test file content"
-
-        file_content_request = FileContentRequest(file_id=file_id, extra_headers=None, extra_body=None)
-
-        # Mock the afile_content method
-        with patch.object(self.handler, "afile_content", new_callable=AsyncMock) as mock_afile_content:
-            mock_response = httpx.Response(
-                status_code=200,
-                content=expected_content,
-                headers={"content-type": "application/octet-stream"},
-                request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
-            )
-            mock_afile_content.return_value = HttpxBinaryResponseContent(response=mock_response)
-
-            # Call the method with _is_async=True
-            result = self.handler.file_content(
-                _is_async=True,
-                file_content_request=file_content_request,
-                api_base="",
-                vertex_credentials=None,
-                vertex_project="test-project",
-                vertex_location="us-central1",
-                timeout=60.0,
-                max_retries=3,
-            )
-
-            # Should return a coroutine since _is_async=True
-            assert asyncio.iscoroutine(result)
-
-            # Await the result
-            final_result = await result
-            assert isinstance(final_result, HttpxBinaryResponseContent)
-            assert final_result.response.content == expected_content
-
-    def test_httpx_response_compatibility(self):
-        """Test that the created HttpxBinaryResponseContent is compatible with expected interface"""
-        # Test the mock response creation logic
-        expected_content = b"test file content"
-        decoded_path = "gs://test-bucket/test-file.txt"
-
-        mock_response = httpx.Response(
-            status_code=200,
-            content=expected_content,
-            headers={"content-type": "application/octet-stream"},
-            request=httpx.Request(method="GET", url=decoded_path),
-        )
-
-        result = HttpxBinaryResponseContent(response=mock_response)
-
-        # Verify the response properties
-        assert result.response.status_code == 200
-        assert result.response.content == expected_content
-        assert result.response.headers["content-type"] == "application/octet-stream"
-
-        # Verify it has the expected interface (matching OpenAI file content response)
-        assert hasattr(result, "response")
-        assert hasattr(result.response, "content")
-        assert hasattr(result.response, "status_code")
-        assert hasattr(result.response, "headers")
diff --git a/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_integration.py b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_integration.py
new file mode 100644
index 00000000000..402c463d134
--- /dev/null
+++ b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_integration.py
@@ -0,0 +1,93 @@
+"""
+Test Vertex AI files integration with main files API
+"""
+
+import pytest
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import litellm
+from litellm.types.llms.openai import HttpxBinaryResponseContent
+
+
+class TestVertexAIFilesIntegration:
+    """Test integration of Vertex AI files with main litellm API"""
+
+
+
+
+    def test_litellm_file_content_vertex_ai_error_cases(self):
+        """Test error handling in vertex_ai file_content"""
+        # Test missing file_id - the VertexAI provider config's
+        # transform_file_content_request should handle empty file_id.
+        # Since the code now goes through base_llm_http_handler, we mock
+        # ProviderConfigManager to return None so it falls through to the
+        # old vertex_ai code path that validates file_id.
+        with patch(
+            "litellm.files.main.ProviderConfigManager.get_provider_files_config",
+            return_value=None,
+        ):
+            with pytest.raises(ValueError, match="file_id is required"):
+                litellm.file_content(
+                    file_id="",  # Empty file_id should cause error
+                    custom_llm_provider="vertex_ai",
+                    vertex_project="test-project",
+                )
+
+    def test_vertex_ai_provider_in_supported_providers_list(self):
+        """Test that vertex_ai is included in supported providers for file_content"""
+        # This test ensures the type annotations and error messages include vertex_ai
+
+        # Test that calling with unsupported provider raises appropriate error
+        with pytest.raises(Exception, match="unsupported_provider' is not a valid LlmProviders") as exc_info:
+            litellm.file_content(
+                file_id="test-file-id",
+                custom_llm_provider="unsupported_provider",  # This should fail
+            )
+
+        # The error message should mention supported providers including vertex_ai
+        error_message = str(exc_info.value)
+        assert "vertex_ai" in error_message or "supported" in error_message.lower()
+
+    @pytest.mark.asyncio
+    async def test_vertex_ai_file_content_with_timeout_and_retries(self):
+        """Test vertex_ai file_content with timeout and retry configuration"""
+        file_id = "gs%3A%2F%2Ftest-bucket%2Ftest-file.txt"
+        expected_content = b"test file content"
+
+        # Create a mock HttpxBinaryResponseContent response
+        import httpx
+
+        mock_response = httpx.Response(
+            status_code=200,
+            content=expected_content,
+            headers={"content-type": "application/octet-stream"},
+            request=httpx.Request(method="GET", url="gs://test-bucket/test-file.txt"),
+        )
+        mock_result = HttpxBinaryResponseContent(response=mock_response)
+
+        # Mock the base_llm_http_handler.retrieve_file_content
+        with patch(
+            "litellm.files.main.base_llm_http_handler.retrieve_file_content",
+            new_callable=MagicMock,
+        ) as mock_retrieve:
+            mock_retrieve.return_value = mock_result
+
+            # Call with custom timeout and max_retries
+            result = await litellm.afile_content(
+                file_id=file_id,
+                custom_llm_provider="vertex_ai",
+                vertex_project="test-project",
+                vertex_location="us-central1",
+                timeout=120,
+                max_retries=5,
+            )
+
+            # Verify the result
+            assert isinstance(result, HttpxBinaryResponseContent)
+            assert result.response.content == expected_content
+
+            # Verify the mock was called
+            mock_retrieve.assert_called_once()
+            # Verify the timeout was passed through
+            call_kwargs = mock_retrieve.call_args.kwargs
+            assert call_kwargs["timeout"] == 120
diff --git a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_streaming.py b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_streaming.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_streaming.py
rename to tests/unit/llms/vertex_ai/files/test_vertex_ai_files_streaming.py
diff --git a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_transformation.py b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_transformation.py
similarity index 95%
rename from tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_transformation.py
rename to tests/unit/llms/vertex_ai/files/test_vertex_ai_files_transformation.py
index 8a249820cbd..48464e79876 100644
--- a/tests/test_litellm/llms/vertex_ai/files/test_vertex_ai_files_transformation.py
+++ b/tests/unit/llms/vertex_ai/files/test_vertex_ai_files_transformation.py
@@ -860,94 +860,6 @@ class TestVertexBatchOutputTransformation:
         binary = b"%PDF-1.4\n%\xc4\xe5\xf2\xe5\xeb\xa7\n" + b"\x00\x01\x02\xff\xfe" * 64
         assert config._try_transform_vertex_batch_output_to_openai(binary) == binary
 
-    def test_streaming_transform_peaks_below_list_pipeline(self, config):
-        """The output transform must stream row-by-row, not build a list of every
-        parsed row and a second list of transformed rows. This guards against a
-        regression to the list pipeline, which peaks at several full copies and
-        OOMs on large result files. The relative comparison cancels shared noise
-        (per-row transform cost, GC timing) and only the list overhead differs.
-        """
-        import gc
-        import tracemalloc
-
-        from litellm.litellm_core_utils.litellm_logging import Logging
-        from litellm.llms.vertex_ai.gemini.vertex_and_google_ai_studio_gemini import (
-            VertexGeminiConfig,
-        )
-
-        def vertex_row(index: int) -> dict:
-            return {
-                "status": "",
-                "processed_time": "2024-11-01T18:13:16.826+00:00",
-                "request": {
-                    "contents": [{"role": "user", "parts": [{"text": "hi"}]}],
-                    "labels": {"litellm_custom_id": f"r-{index}"},
-                },
-                "response": {
-                    "candidates": [
-                        {
-                            "content": {
-                                "parts": [{"text": "hello " * 20}],
-                                "role": "model",
-                            },
-                            "finishReason": "STOP",
-                        }
-                    ],
-                    "modelVersion": "gemini-2.0-flash-001",
-                    "usageMetadata": {
-                        "promptTokenCount": 10,
-                        "candidatesTokenCount": 20,
-                        "totalTokenCount": 30,
-                    },
-                },
-            }
-
-        content = ("\n".join(json.dumps(vertex_row(i)) for i in range(4000))).encode("utf-8")
-
-        def list_pipeline() -> bytes:
-            gemini_config = VertexGeminiConfig()
-            logging_obj = Logging(
-                model="",
-                messages=[],
-                stream=False,
-                call_type="batch_transform",
-                start_time=0.1,
-                litellm_call_id="",
-                function_id="",
-            )
-            logging_obj.optional_params = {}
-            mock_response = httpx.Response(
-                status_code=200,
-                headers={"content-type": "application/json"},
-                request=httpx.Request("POST", "https://example.com"),
-            )
-            rows = content.decode("utf-8").strip().split("\n")
-            transformed = [
-                json.dumps(
-                    config._transform_single_vertex_batch_output_to_openai(
-                        json.loads(row), gemini_config, logging_obj, mock_response
-                    )
-                )
-                for row in rows
-            ]
-            return "\n".join(transformed).encode("utf-8")
-
-        def peak_of(fn) -> int:
-            gc.collect()
-            tracemalloc.start()
-            try:
-                fn()
-                return tracemalloc.get_traced_memory()[1]
-            finally:
-                tracemalloc.stop()
-
-        streaming_peak = peak_of(lambda: config._try_transform_vertex_batch_output_to_openai(content))
-        list_peak = peak_of(list_pipeline)
-
-        assert streaming_peak < list_peak * 0.75, (
-            f"streaming peak {streaming_peak} is not a clear win over the list "
-            f"pipeline {list_peak} (ratio {streaming_peak / list_peak:.2f})"
-        )
 
 
 class TestTryTransformDoesNotMutateCallerLoggingObj:
diff --git a/tests/unit/llms/vertex_ai/gemini_embeddings/__init__.py b/tests/unit/llms/vertex_ai/gemini_embeddings/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/gemini_embeddings/test_batch_embed_content_transformation.py b/tests/unit/llms/vertex_ai/gemini_embeddings/test_batch_embed_content_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/gemini_embeddings/test_batch_embed_content_transformation.py
rename to tests/unit/llms/vertex_ai/gemini_embeddings/test_batch_embed_content_transformation.py
diff --git a/tests/unit/llms/vertex_ai/image_edit/__init__.py b/tests/unit/llms/vertex_ai/image_edit/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/image_edit/test_vertex_ai_image_edit_transformation.py b/tests/unit/llms/vertex_ai/image_edit/test_vertex_ai_image_edit_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/image_edit/test_vertex_ai_image_edit_transformation.py
rename to tests/unit/llms/vertex_ai/image_edit/test_vertex_ai_image_edit_transformation.py
diff --git a/tests/unit/llms/vertex_ai/interactions/__init__.py b/tests/unit/llms/vertex_ai/interactions/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/interactions/test_vertex_ai_interactions_transformation.py b/tests/unit/llms/vertex_ai/interactions/test_vertex_ai_interactions_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/interactions/test_vertex_ai_interactions_transformation.py
rename to tests/unit/llms/vertex_ai/interactions/test_vertex_ai_interactions_transformation.py
diff --git a/tests/unit/llms/vertex_ai/multimodal_embeddings/__init__.py b/tests/unit/llms/vertex_ai/multimodal_embeddings/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/multimodal_embeddings/test_vertex_ai_multimodal_embedding_transformation.py b/tests/unit/llms/vertex_ai/multimodal_embeddings/test_vertex_ai_multimodal_embedding_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/multimodal_embeddings/test_vertex_ai_multimodal_embedding_transformation.py
rename to tests/unit/llms/vertex_ai/multimodal_embeddings/test_vertex_ai_multimodal_embedding_transformation.py
diff --git a/tests/unit/llms/vertex_ai/realtime/__init__.py b/tests/unit/llms/vertex_ai/realtime/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py b/tests/unit/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py
similarity index 92%
rename from tests/test_litellm/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py
rename to tests/unit/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py
index d4cf58bc0b4..a71ef21d997 100644
--- a/tests/test_litellm/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py
+++ b/tests/unit/llms/vertex_ai/realtime/test_vertex_ai_realtime_transformation.py
@@ -387,50 +387,6 @@ def test_vertex_does_not_warn_when_dropping_non_guardrail_session_update(caplog)
     )
 
 
-@pytest.mark.asyncio
-async def test_async_realtime_does_not_forward_client_query_params_to_vertex_backend(
-    monkeypatch,
-):
-    """Regression: forwarding client ?model=/?intent= to the Vertex Live WSS URL causes 1007 errors.
-
-    Exercises ``async_realtime`` end-to-end so that re-adding ``_append_query_params``
-    (the reverted bug) would push ``model=``/``intent=`` onto the backend URL and fail here.
-    """
-    import websockets
-
-    from litellm.llms.custom_httpx.llm_http_handler import BaseLLMHTTPHandler
-
-    cfg = VertexAIRealtimeConfig(
-        access_token="tok", project="my-proj", location="us-central1"
-    )
-
-    captured: dict = {}
-
-    def fake_connect(url, *args, **kwargs):
-        captured["url"] = url
-        raise RuntimeError("stop before establishing the backend connection")
-
-    monkeypatch.setattr(websockets, "connect", fake_connect)
-
-    try:
-        await BaseLLMHTTPHandler().async_realtime(
-            model="gemini-live-2.5-flash-preview-native-audio-09-2025",
-            websocket=AsyncMock(),
-            logging_obj=MagicMock(),
-            provider_config=cfg,
-            headers={},
-            query_params={
-                "model": "gemini-live-2.5-flash-preview-native-audio-09-2025",
-                "intent": "chat",
-            },
-        )
-    except (RuntimeError, Exception):
-        pass
-
-    assert "url" in captured, "websockets.connect was never called"
-    assert "?" not in captured["url"]
-    assert "model=" not in captured["url"]
-    assert "intent=" not in captured["url"]
 
 
 def test_vertex_function_call_output_omits_id():
diff --git a/tests/unit/llms/vertex_ai/text_to_speech/__init__.py b/tests/unit/llms/vertex_ai/text_to_speech/__init__.py
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/tests/test_litellm/llms/vertex_ai/text_to_speech/test_transformation.py b/tests/unit/llms/vertex_ai/text_to_speech/test_transformation.py
similarity index 100%
rename from tests/test_litellm/llms/vertex_ai/text_to_speech/test_transformation.py
rename to tests/unit/llms/vertex_ai/text_to_speech/test_transformation.py