Fix: Vertex AI Gemini labels field provider-aware filtering (#14563)

* Add comprehensive tests for Vertex AI Gemini labels provider filtering

- Test Google GenAI endpoints exclude labels even when explicitly provided
- Test Vertex AI endpoints include labels when provided
- Cover provider detection logic for different endpoint URLs
- Verify metadata-to-labels conversion only happens for Vertex AI
- Ensure edge cases are handled properly (null/empty api_base)

* Fix Vertex AI Gemini labels field provider-aware filtering

- Add _is_google_genai_endpoint() function to detect Google GenAI vs Vertex AI endpoints
- Update _transform_request_body() to accept api_base parameter
- Only include labels field for Vertex AI endpoints (not Google GenAI)
- Pass api_base through sync/async transform functions
- Maintain backward compatibility with existing usage
- Fixes issue where Google GenAI requests failed with unsupported labels field

* Refactor labels filtering to use custom_llm_provider instead of URL parsing

Replace URL-based endpoint detection with custom_llm_provider parameter
checking for cleaner, more reliable provider identification.

Changes:
- Remove _is_google_genai_endpoint() helper function
- Update labels condition to use custom_llm_provider != "gemini"
- Remove api_base parameter from _transform_request_body()
- Simplify sync/async transform function signatures
- Update tests to reflect new parameter structure
- Remove obsolete test_provider_detection test

This approach aligns with existing codebase patterns where
custom_llm_provider="gemini" identifies Google AI Studio endpoints
that don't support labels, while vertex_ai/vertex_ai_beta identify
Vertex AI endpoints that do support labels.

* Use LlmProviders.GEMINI constant instead of hardcoded string

litellm/llms/vertex_ai/gemini/transformation.py

@@ -28,6 +28,7 @@ from litellm.types.files import (
     get_file_type_from_extension,
     is_gemini_1_5_accepted_file_type,
 )
+from litellm.types.utils import LlmProviders
 from litellm.types.llms.openai import (
     AllMessageValues,
     ChatCompletionAssistantMessage,
@@ -492,7 +493,8 @@ def _transform_request_body(
             data["generationConfig"] = generation_config
         if cached_content is not None:
             data["cachedContent"] = cached_content
-        if labels is not None:
+        # Only add labels for Vertex AI endpoints (not Google GenAI/AI Studio) and only if non-empty
+        if labels and custom_llm_provider != LlmProviders.GEMINI:
             data["labels"] = labels
     except Exception as e:
         raise e
@@ -647,3 +649,5 @@ def _transform_system_message(
         return SystemInstructions(parts=system_content_blocks), messages
 
     return None, messages
+
+

tests/test_litellm/llms/vertex_ai/gemini/test_vertex_ai_gemini_transformation.py

@@ -1,4 +1,7 @@
-from litellm.llms.vertex_ai.gemini.transformation import check_if_part_exists_in_parts
+from litellm.llms.vertex_ai.gemini.transformation import (
+    check_if_part_exists_in_parts,
+    _transform_request_body,
+)
 
 
 def test_check_if_part_exists_in_parts():
@@ -73,3 +76,82 @@ def test_check_if_part_exists_in_parts_camel_case_snake_case():
     }
 
     assert check_if_part_exists_in_parts(parts_mixed, part_mixed_casing)
+
+
+# Tests for issue #14556: Labels field provider-aware filtering
+def test_google_genai_excludes_labels():
+    """Test that Google GenAI/AI Studio endpoints exclude labels when custom_llm_provider='gemini'"""
+    messages = [{"role": "user", "content": "test"}]
+    optional_params = {"labels": {"project": "test", "team": "ai"}}
+    litellm_params = {}
+
+    result = _transform_request_body(
+        messages=messages,
+        model="gemini-2.5-pro",
+        optional_params=optional_params,
+        custom_llm_provider="gemini",
+        litellm_params=litellm_params,
+        cached_content=None,
+    )
+
+    # Google GenAI/AI Studio should NOT include labels
+    assert "labels" not in result
+    assert "contents" in result
+
+
+def test_vertex_ai_includes_labels():
+    """Test that Vertex AI endpoints include labels when custom_llm_provider='vertex_ai'"""
+    messages = [{"role": "user", "content": "test"}]
+    optional_params = {"labels": {"project": "test", "team": "ai"}}
+    litellm_params = {}
+
+    result = _transform_request_body(
+        messages=messages,
+        model="gemini-2.5-pro",
+        optional_params=optional_params,
+        custom_llm_provider="vertex_ai",
+        litellm_params=litellm_params,
+        cached_content=None,
+    )
+
+    # Vertex AI SHOULD include labels
+    assert "labels" in result
+    assert result["labels"] == {"project": "test", "team": "ai"}
+
+
+
+def test_metadata_to_labels_vertex_only():
+    """Test that metadata->labels conversion only happens for Vertex AI"""
+    messages = [{"role": "user", "content": "test"}]
+    optional_params = {}
+    litellm_params = {
+        "metadata": {
+            "requester_metadata": {
+                "user": "john_doe",
+                "project": "test-project"
+            }
+        }
+    }
+
+    # Google GenAI/AI Studio should not include labels from metadata
+    result = _transform_request_body(
+        messages=messages,
+        model="gemini-2.5-pro",
+        optional_params=optional_params.copy(),
+        custom_llm_provider="gemini",
+        litellm_params=litellm_params.copy(),
+        cached_content=None,
+    )
+    assert "labels" not in result
+
+    # Vertex AI should include labels from metadata
+    result = _transform_request_body(
+        messages=messages,
+        model="gemini-2.5-pro",
+        optional_params=optional_params.copy(),
+        custom_llm_provider="vertex_ai",
+        litellm_params=litellm_params.copy(),
+        cached_content=None,
+    )
+    assert "labels" in result
+    assert result["labels"] == {"user": "john_doe", "project": "test-project"}