diff --git a/CHANGELOG.md b/CHANGELOG.md
index e54fff1dd..53df5c1dc 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -9,6 +9,7 @@
 ### Changed
 
 - (**breaking**) `AnthropicLLM.supports_structured_output` is now `True`. As a result, `SchemaFromTextExtractor` and `LLMEntityRelationExtractor` (and `SimpleKGPipeline`, which enables structured output automatically when the LLM supports it) now use structured output by default with `AnthropicLLM`. This requires a Claude 4.5+ model (e.g. `claude-sonnet-4-5`); using `AnthropicLLM` with an older Claude model in these components will now raise an error where it previously worked. To keep the previous behavior, use a Claude 4.5+ model, or construct `LLMEntityRelationExtractor` / `SchemaFromTextExtractor` directly with `use_structured_output=False`.
+- (**breaking**) `GeminiLLM.supports_structured_output` is now `True`. `GeminiLLM` already implemented structured-output handling via `response_schema`/`response_mime_type` but never declared the capability flag, so components gating on it saw `False`. As a result, `SchemaFromTextExtractor` and `LLMEntityRelationExtractor` (and `SimpleKGPipeline`, which enables structured output automatically when the LLM supports it) now use structured output by default with `GeminiLLM`. To keep the previous behavior, construct `LLMEntityRelationExtractor` / `SchemaFromTextExtractor` directly with `use_structured_output=False`.
 
 ## 1.18.0
 
diff --git a/src/neo4j_graphrag/llm/google_genai_llm.py b/src/neo4j_graphrag/llm/google_genai_llm.py
index 52cee38fd..eb9e930e3 100644
--- a/src/neo4j_graphrag/llm/google_genai_llm.py
+++ b/src/neo4j_graphrag/llm/google_genai_llm.py
@@ -72,6 +72,8 @@ class GeminiLLM(LLMInterface, LLMInterfaceV2):
         **kwargs (Any): Arguments passed to the genai.Client.
     """
 
+    supports_structured_output: bool = True
+
     def __init__(
         self,
         model_name: str = "gemini-2.0-flash",
diff --git a/tests/unit/llm/test_google_genai_llm.py b/tests/unit/llm/test_google_genai_llm.py
index 997d985a1..a54cc49f1 100644
--- a/tests/unit/llm/test_google_genai_llm.py
+++ b/tests/unit/llm/test_google_genai_llm.py
@@ -134,3 +134,12 @@ def test_gemini_invoke_error(mock_genai: Tuple[MagicMock, MagicMock]) -> None:
     llm = GeminiLLM("gemini-2.0-flash")
     with pytest.raises(LLMGenerationError):
         llm.invoke("hello")
+
+
+def test_gemini_llm_supports_structured_output(
+    mock_genai: Tuple[MagicMock, MagicMock],
+) -> None:
+    """GeminiLLM implements structured output (response_schema/response_mime_type),
+    so it must declare the capability flag components gate on."""
+    llm = GeminiLLM("gemini-2.0-flash")
+    assert llm.supports_structured_output is True