fix(models): forward Anthropic generation config

Author: MukundaKatta

src/google/adk/models/anthropic_llm.py

@@ -377,6 +377,22 @@ def _resolve_model_name(self, model: Optional[str]) -> str:
         return match.group(1)
     return model
 
+  def _get_generation_kwargs(
+      self, llm_request: LlmRequest
+  ) -> dict[str, Any]:
+    generation_kwargs: dict[str, Any] = {}
+
+    if llm_request.config.temperature is not None:
+      generation_kwargs["temperature"] = llm_request.config.temperature
+    if llm_request.config.top_p is not None:
+      generation_kwargs["top_p"] = llm_request.config.top_p
+    if llm_request.config.top_k is not None:
+      generation_kwargs["top_k"] = llm_request.config.top_k
+    if llm_request.config.stop_sequences:
+      generation_kwargs["stop_sequences"] = llm_request.config.stop_sequences
+
+    return generation_kwargs
+
   @override
   async def generate_content_async(
       self, llm_request: LlmRequest, stream: bool = False
@@ -401,6 +417,7 @@ async def generate_content_async(
         if llm_request.tools_dict
         else NOT_GIVEN
     )
+    generation_kwargs = self._get_generation_kwargs(llm_request)
 
     if not stream:
       message = await self._anthropic_client.messages.create(
@@ -410,6 +427,7 @@ async def generate_content_async(
           tools=tools,
           tool_choice=tool_choice,
           max_tokens=self.max_tokens,
+          **generation_kwargs,
       )
       yield message_to_generate_content_response(message)
     else:
@@ -439,6 +457,7 @@ async def _generate_content_streaming(
         tool_choice=tool_choice,
         max_tokens=self.max_tokens,
         stream=True,
+        **self._get_generation_kwargs(llm_request),
     )
 
     # Track content blocks being built during streaming.

tests/unittests/models/test_anthropic_llm.py

@@ -1350,3 +1350,108 @@ async def test_non_streaming_does_not_pass_stream_param():
   mock_client.messages.create.assert_called_once()
   _, kwargs = mock_client.messages.create.call_args
   assert "stream" not in kwargs
+
+
+@pytest.mark.asyncio
+async def test_non_streaming_forwards_generation_config_kwargs():
+  llm = AnthropicLlm(model="claude-sonnet-4-20250514")
+
+  mock_message = anthropic_types.Message(
+      id="msg_test",
+      content=[
+          anthropic_types.TextBlock(text="Hello!", type="text", citations=None)
+      ],
+      model="claude-sonnet-4-20250514",
+      role="assistant",
+      stop_reason="end_turn",
+      stop_sequence=None,
+      type="message",
+      usage=anthropic_types.Usage(
+          input_tokens=5,
+          output_tokens=2,
+          cache_creation_input_tokens=0,
+          cache_read_input_tokens=0,
+          server_tool_use=None,
+          service_tier=None,
+      ),
+  )
+
+  mock_client = MagicMock()
+  mock_client.messages.create = AsyncMock(return_value=mock_message)
+
+  llm_request = LlmRequest(
+      model="claude-sonnet-4-20250514",
+      contents=[Content(role="user", parts=[Part.from_text(text="Hi")])],
+      config=types.GenerateContentConfig(
+          system_instruction="Test",
+          temperature=0.0,
+          top_p=0.8,
+          top_k=12,
+          stop_sequences=["DONE"],
+      ),
+  )
+
+  with mock.patch.object(llm, "_anthropic_client", mock_client):
+    _ = [r async for r in llm.generate_content_async(llm_request, stream=False)]
+
+  _, kwargs = mock_client.messages.create.call_args
+  assert kwargs["temperature"] == 0.0
+  assert kwargs["top_p"] == 0.8
+  assert kwargs["top_k"] == 12
+  assert kwargs["stop_sequences"] == ["DONE"]
+
+
+@pytest.mark.asyncio
+async def test_streaming_forwards_generation_config_kwargs():
+  llm = AnthropicLlm(model="claude-sonnet-4-20250514")
+
+  events = [
+      MagicMock(
+          type="message_start",
+          message=MagicMock(usage=MagicMock(input_tokens=5, output_tokens=0)),
+      ),
+      MagicMock(
+          type="content_block_start",
+          index=0,
+          content_block=anthropic_types.TextBlock(text="", type="text"),
+      ),
+      MagicMock(
+          type="content_block_delta",
+          index=0,
+          delta=anthropic_types.TextDelta(text="Hi", type="text_delta"),
+      ),
+      MagicMock(type="content_block_stop", index=0),
+      MagicMock(
+          type="message_delta",
+          delta=MagicMock(stop_reason="end_turn"),
+          usage=MagicMock(output_tokens=1),
+      ),
+      MagicMock(type="message_stop"),
+  ]
+
+  mock_client = MagicMock()
+  mock_client.messages.create = AsyncMock(
+      return_value=_make_mock_stream_events(events)
+  )
+
+  llm_request = LlmRequest(
+      model="claude-sonnet-4-20250514",
+      contents=[Content(role="user", parts=[Part.from_text(text="Hi")])],
+      config=types.GenerateContentConfig(
+          system_instruction="Test",
+          temperature=0.2,
+          top_p=0.7,
+          top_k=8,
+          stop_sequences=["STOP_HERE"],
+      ),
+  )
+
+  with mock.patch.object(llm, "_anthropic_client", mock_client):
+    _ = [r async for r in llm.generate_content_async(llm_request, stream=True)]
+
+  _, kwargs = mock_client.messages.create.call_args
+  assert kwargs["stream"] is True
+  assert kwargs["temperature"] == 0.2
+  assert kwargs["top_p"] == 0.7
+  assert kwargs["top_k"] == 8
+  assert kwargs["stop_sequences"] == ["STOP_HERE"]